WK 5 Combined Psychological Therapies PDF

Module: Psychological Foundations of Mental Health Week 5 Psychological therapies: from behaviour modiﬁcation to behaviour therapy Topic 1 The ﬁrst wave - behavioural psychotherapy - Part 1 of 3 Professor Richard Brown Department of Psychology, King’s College London Lecture transcript Slide 4 The potential for classical and operant conditioning to alter human behaviour was evident almost from the outset of behaviourism. Although neither Pavlov, Skinner, nor Watson conducted clinical research or sought to treat people with mental health problems, their work provided the essential foundation for a whole new behavioural approach to therapy. This was to challenge and eventually replace the predominant approach at the time, based on the psychoanalytic theories of Freud, Jung, and others. Behavioural or ﬁrst wave psychotherapy arose directly from the models and evidence of behaviourism and learning theory. In this ﬁrst topic, we will trace and describe how behavioural approaches developed on both sides of the Atlantic from the middle of the 20th century. These approaches have often been called the American and the British traditions. The names are somewhat misleading, as their developers and proponents, particularly the so- called British tradition, came from all areas of the world. However, there are some important distinguishing features between the two approaches. First, in how their methods drew on different aspects of behavioural theory and evidence. Second, the settings for which they were developed and used. And third, the nature of the problems that they sought to address. Slide 5 Let’s look at the US tradition ﬁrst. This was most strongly and directly inﬂuenced by the work of Skinner. It sought to apply operant methods to change behaviour, rather than treat emotional disturbance. For this reason, it is commonly known as behaviour modiﬁcation. Operant behavioural approaches were most widely applied for use in psychiatric institutions or asylums, where people were committed for months or even years, as well as institutions for children and adults with severe learning disabilities or older adults with dementia. For those unfortunate to be placed in such institutions, behavioural problems were common. Some risked harm to the individual themselves, such as self-injury or eating non-food objects, aggression that might harm others, or behaviour that was seen as disruptive, such as soiling, refusal to wear clothes, and prolonged screaming. Today, these are typically referred to by the term “challenging behaviours.” Transcripts by 3Playmedia Week © King’s College London 2018 1. For psychologists at the time, these were seen as ideally suited to the principles and methods of operant conditioning, using positive and negative reinforcers to humanely increase the frequency of a positive or desired behaviour and so reduce the negative behaviours. Slide 6 Operant approaches were directly based on the three-term model that we covered in week one, the discriminative stimulus, the behaviour, and the reinforcer. In behaviour modiﬁcation, they were more typically deﬁned as the antecedent stimulus, the behaviour, and the consequence or outcome. This outcome was presumed to be the reinforcing event that maintained the behaviour. This three-term model is commonly called the ABC model. Outside of the controlled laboratory situation, in the real world, especially somewhere as complex as an institutional setting for people with severe mental illness, identifying the speciﬁc stimulus that provokes the behaviour and the speciﬁc consequences that maintain it was far from easy. Without a good understanding of these relationships, operant methods would struggle to ﬁnd the best ways to produce behaviour change. To address this need, the techniques of applied behavioural analysis were developed and are still used today. Today they are also called functional analysis or functional assessment. The term “functional” is used because we assume that the behaviour in question has a speciﬁc function or purpose for that individual, to either gain a positive outcome or avoid a negative one. Such a purpose may be highly speciﬁc and individual. However, they tend to fall into two broad categories, either to obtain or access something positive for the individual, or to escape or avoid something negative. Let’s look at behaviours that can be reinforced by access to something positive. First, behaviour may serve the function of gaining attention from other people, whether from staff or other residents. In a busy setting and for people who may have poor communication skills, a disruptive behaviour, such as falling from a chair, may quickly bring a medical staff running and provides some one-to-one attention. Second, a behaviour may result in some tangible reward, such as food. A person who screams loudly until they get their meal may often ﬁnd that they get fed quicker than others. Finally, the behaviour may provide sensory stimulation. Behaviours such as rocking, shouting, or self-harm may provide stimulation in impoverished environments or distraction from negative emotions or physical discomfort. In terms of escape or avoidance, the behaviour may be to allow them to get away from situations that they ﬁnd uncomfortable or distressing. While some individuals may want attention, others ﬁnd social interaction aversive and anxiety provoking. Disrupting a social situation may result in the person being sent away or not be included in such events in the future. Avoidance of an unpleasant or demanding task or situation may also be the purpose served by disruptive behaviour. For example, someone who has to do chores, such as clearing tables, may be taken off that duty if they repeatedly drop the plates. Slide 7 A number of screening tools have been developed to quickly assess possible situations and reinforcers, prior to a full functional analysis. One recent example, shown here, is the functional assessment screening tool developed by Brian Iwata and colleagues at the University of Florida for use with children with autism in the home, school, or other settings. The brief, 16 item questionnaire is shown here. It collects information from a number of people or Transcripts by 3Playmedia Week © King’s College London 2018 2. informants who regularly interact with the adult or child in question, whether health professional, family member, teacher, and so on. Pause if you want to look over the scale to see how it is constructed and the questions asked. Think about what ABC contingencies it is exploring. Remember, this is intended as a screening tool to guide further functional assessment. It provides an easy means of doing a preliminary, indirect analysis, based on reports of informants, rather than direct observation. Slide 8 A full behavioural or functional analysis is a systematic approach that is typically based on careful, direct observation over a period of time, either by staff in a hospital or care setting, teachers in a school, or family members at home. It requires record keeping, using so-called ABC charts. Using such tools accurately and reliably requires careful training. The charts involve the assessor noting the precise characteristics and details of the observed behaviour and then recording the situation that immediately preceded it, the antecedents, and what happened immediately after, the consequences. The latter might include how others reacted and any changes in the behaviour of the individual. Note there is no interpretation of the situation, simply a reporting of what is observed. Slide 9 These two rather straightforward illustrations show how a problem behaviour, disruptiveness at the meal table in the ﬁrst and self-harm in the second, followed straight on from situations in which the patient, X, was made either to wait for their meal or denied the opportunity to continue watching television. The consequences in both cases was a member of staff giving X what they wanted, which led to the challenging behaviour ceasing. It’s fairly evident from these simple examples that the staff member was inadvertently reinforcing the undesired behaviour, therefore increasing the probability that it would be repeated in the future. We should note that such simple contingencies are not restricted to clinical settings. The practice of supermarkets placing sweets near check-out aisles has led to many parents reinforcing temper tantrums in their toddlers. They put up with the screaming for as long as they can, before their patience or embarrassment causes them to crack and buy a packet of sweets. Not surprisingly, the child learns quickly how to get sweets in the future. The use of partial reinforcement makes it even worse. Slide 10 ABC analyses can lead to a range of different approaches to reduce the undesired behaviour. These fall into three main classes, those that address the antecedents, the behaviour itself, or the consequences. First, let’s look at an approach based on the antecedents. Let’s imagine that an ABC analysis identiﬁes that an individual’s challenging behaviour typically follows high levels of noise, a potentially aversive and distressing stimulus. The behaviour leads to them being removed from the situation or others making the noise asked to leave. In such a situation, reducing the individual’s exposure to noise at key times, such as meals, may be all that is needed. This might also involve serving the individual at a quieter table or at a quieter time, or allowing them to move if they want to. If a reduction in the behaviour is observed as a result, the analysis would seem to be supported and the intervention successful. This approach may be much simpler and easier to implement and more humane than one aimed at increasing the individual’s ability to tolerate loud noise. Transcripts by 3Playmedia Week © King’s College London 2018 3. Next, let’s look at behavior-based approaches. These assume that a challenging behaviour may arise to achieve a purpose because the individual lacks alternative ways to achieve it. To take the mealtime noise example again, an individual who lacks effective verbal communication skills may be unable to ask for what they want and so ﬁnd other ways to get it. This can be addressed by providing new skills, in this case, new and more effective ways of communicating. This may involve providing simple tools to allow non-verbal communication. Such an approach can broaden their behavioural repertoire, increasing their ability to communicate their needs and wishes. Such skills-based approaches are often more costly in time and effort to implement, but pay dividends in that the new skill behaviour is available for use by the individual in other situations. Skills-based approaches are also applicable in non-clinical settings. Episodes of aggression in day- to-day life can result where an individual, who can otherwise communicate effectively, lacking the repertoire of appropriate interpersonal skills to manage difficult interactions with other people. A programme of assertiveness or social skills training can be effective in reducing the unwanted behaviour. This is still the application of a behavioural skills-based approach within the ABC model. Finally, let’s consider strategies based on the consequences, the outcomes, essentially, what reinforcement is being obtained to maintain the problem behaviour. Having identiﬁed a possible candidate contingency from an ABC analysis, the easiest option is simply to break the link between the behaviour and the reinforcing outcome. For example, if episodes of screaming seem to result in increased attention, it suggests that gaining attention may be the purpose of the behaviour. Evidence and operant conditioning studies suggest that the screaming will eventually stop or extinguish if the attention is no longer given contingently on the behaviour, in other words, that it is ignored. While this is likely to work, it can be hard to apply and neglects the underlying purpose of the behaviour. First, some behaviours cannot safely or practically be ignored. Even if they are for most of the time, a single lapse by a parent, teacher, or member of staff will undo the hard work done so far. Think back to the reinforcement schedules that you learned about in week one. If a person gets reinforced every 10th time that they behave in a certain way, they are effectively on a partial reinforcement schedule. We learned how in rats and pigeons, this led to increased resistance to extinction. The same applies to us as human beings. When a parent eventually cracks and gives into the screams of their toddler, they are doing the worst possible thing for changing the behaviour, even if we all understand why they did it. Indeed, if we are going to use an extinction approach, the best way to do it is to give the reinforcement immediately and consistently every time the behaviour happens for a period, before we remove the reinforcement completely, in other words, setting up a continuous reinforcement schedule, one that we know leads to the fastest extinction. However, because extinction is so hard to put into practice, the approach is more usually not simply to ignore the behaviour, but to provide the reinforcement non-contingently or more often, at a time when the unwanted behaviour is not occurring. To take the attention seeking example again, minimal required attention may still be given to reduce disruption, but is provided more often and more positively at other times, when the individual is quiet. This serves both to break the link between the unwanted behaviour and the outcome, and reinforces a more helpful behaviour. All of this accepts that the behaviour is designed for a perfectly valid and positive purpose, to gain attention and social interaction. There is no attempt to deny that purpose, simply to alter the ways in which the individual achieves it. Transcripts by 3Playmedia Week © King’s College London 2018 4. These examples give you a ﬂavour of the types of approaches that can be taken on an individual basis to both reduce challenging behaviours, while typically, at the same time, increasing alternative behaviours that provide valued needs for the individual. Slide 11 Let’s look at an example of how a simple ABC approach can be used effectively in practice in dementia care settings, with staff who have only a basic knowledge of functional analysis and behaviour change techniques. A range of challenging behaviours can be observed in people with dementia, typically termed “agitation,” including aggression, swearing, screaming, refusal to take food or liquids. These can become increasingly common as the condition progresses. The onset of persistent agitation is often the trigger that causes some family carers to place their relative in a formal care setting. As a result, such behaviours are common in dementia care homes, and their management is a clinical priority. Although sedating drugs have been widely used in the past, there is an ever-growing focus on the use of behavioural approaches, including functional analysis. These have led to the development of the simple tool kit approach that care home staff can be easily trained to deliver. One such tool kit has been developed by the American psychiatrist Jessica Cohen-Mansﬁeld. This forms part of a comprehensive assessment and management programme, involving individualised ABC assessments and individualised treatment. This is called the Treatment Routes for Exploring Agitation programme. Here’s an example for one particular behaviour, verbal agitation. Rather than using a blank ABC chart, it provides the staff with the main situations in which the behaviour is observed and so suggests possible causes. For each, an intervention is suggested that might reduce the behaviour by introducing new positive consequences or take away negative ones. Common antecedents may include pain and discomfort that cannot be communicated effectively, the need for social contact, and boredom. Targeted interventions arising from these include reducing the pain or discomfort, providing social contact, or ﬁnding meaningful activities to reduce the boredom. An important point about this approaches is that it places the needs of the person at the centre. The aim is less to manage the problem behaviour for the staff’s beneﬁt and more to address the individual’s basic needs. Slide 12 The TREA approach has been shown to be effective at reducing agitation in clinical trials. In this randomised control trial, 12 nursing homes took part, with half allocated to the TREA approach and half to continue with their standard care. This design is known as a cluster randomised trial. This means that the centres are randomised, not individual patients within them. The intervention involved 10 days of individualised treatment for residents at the care homes, during the four hours of greatest observed agitation. Slide 13 The ﬁgure here shows the mean agitation recorded for residents across all of the homes in the two groups, before and after the intervention. Agitation was measured using a standard rating scale, completed by an observer. The results show greater reduction in mean agitation in homes using the TREA method than those using standard care. Also shown in the table below is improvement in indicators of patient pleasure and interest, although no change was noted in negative affect or mood. Other research has shown that Transcripts by 3Playmedia Week © King’s College London 2018 5. sustained use of such person-centered approaches have a range of other beneﬁts, such as reduced use of tranquillising medication and lower levels of burnout in care staff. Slide 14 In some situations, obvious operant approaches struggle, as there does not seem to be an effective reinforcer that can be applied. The person may refuse typical reinforcers or, if accepted, they may be ineffective, because they are not valued. Remember, reinforcers can only be considered such if behaviour changes as a result of their pairing with a particular response. While many primary and secondary reinforcers are universal, some people with severe mental health problems or with cognitive impairment may not respond with a change in behaviour, perhaps because the consequence is not valued. A solution that is sometimes used in such situations is to apply work undertaken in animals by the behaviourist David Premack and is called Premack’s principle. This proposes that a behaviour that is chosen frequently by an animal or human is itself reinforcing. It follows that a frequently chosen behaviour can be used as a reinforcer to alter another behaviour, particularly one that is chosen less frequently. Although sounding technical, the basic principle is widely used without us realising it. A parent who says, you have to eat your cabbage before you can watch TV, is using a high frequency preferred behaviour, television watching, to increase a low frequency and less preferred behaviour, eating cabbage. This does not depend on making any assumptions that TV watching itself is more rewarding. Slide 15 Let’s look at an example of this put into practice in a clinical setting. This study was by William Mitchell and Bertram Stoffelmayr, published in 1975. Extreme inactivity is one way in which schizophrenia can manifest, particularly in institutional settings of the time. Patients would sit for long periods of time, often resisting efforts for them to engage in other activities. These are examples of so-called negative symptoms in schizophrenia. This study used Premack’s principle in a psychiatric hospital, with a group of patients who were failing to engage in the activities offered. Typical of the time, these were very low demand, mundane activities that fall under the rather optimistic name of industrial therapy. Today, we would argue that changing the activity was more useful than changing the patients’ engagement with it. Nevertheless, this study illustrates how Premack’s principle can be used. The ﬁgure on the left shows the work activity observed in a group of patients over six, 30 minute sessions. While activity was variable in most patients, in three, shown at the bottom, activity was very low. Two of these were identiﬁed, and a behavioural programme devised to increase their engagement in the work task. The high frequency behaviour identiﬁed was sitting, something that these patients did most of the time if permitted. The behavioural intervention involved them being required to stand while at the workplace and being allowed to sit only if performing the task, in other words, working. A process of active shaping was used, followed by a period of time when it was observed. The intervention was evaluated using what is known as an AB design. A control, non-intervention condition is assessed ﬁrst, in this case, simple verbal instructions to work, followed by the intervention condition, B. In this case, the design was further reﬁned by reversing the A condition before reinstating condition B. Such designs use the patient’s own behaviour as a control for the intervention. If behaviour changes systematically with the intervention, there is greater conﬁdence that it is effective and that an effective reinforcer has been identiﬁed. Transcripts by 3Playmedia Week © King’s College London 2018 6. What we see here is the change in behaviour over the time in the two patients. We see that activity was initially completely absent, but increased signiﬁcantly during the shaping period and continued so long as they were able to sit only when working. When this contingency was no longer in place, work activity stopped, and they sat non-contingently. Behaviour quickly resumed once the sitting rule was put in place. We can look back at this example and feel uncomfortable about the manipulation of patients being made to do a fairly purposeless activity that probably had no therapeutic value. The potentially coercive nature of such behaviour modiﬁcation approaches has led to concern about their use, particularly with vulnerable adults and children. However, Premack’s principle remains a useful approach when obvious reinforcers are hard to ﬁnd and are still used in more client-centered approaches. Slide 16 One of the most systematic applications of operant principles came with the development of so- called token economy methods. This is a form of contingency management based on the principle of secondary reinforcement. You will remember from week one that some outcomes have reinforcing value through their association with others, often primary or tangible reinforcers. We considered how money, itself just a piece of paper or metal or even ﬁgures on a bank statement, have no intrinsic value, but can motivate and reinforce behaviour. Secondary reinforcers can therefore be considered tokens, something that allows us to acquire an outcome and so achieve a valued purpose, typically at a later time. Using tokens as reinforces offers a number of potential advantages. The ﬁrst is that they are easier to control and manage than other reinforcers. Providing a reinforcer at the time of the behaviour leads to the quickest learning. However, it is often neither possible nor desirable to use tangible reinforcers in this way. For example, a member of staff could not immediately provide attention or allow access to a quiet area immediately a desired behaviour occurred. Tokens, therefore, allow any gap between the time of the behaviour and the ultimate reinforcing outcome to be effectively bridged. Second, the reinforcement is easily scalable. The value of the token can be adjusted as part of the process of shaping the desired behaviour. In other words, the exchange rate can be easily modiﬁed. As part of this, tokens can be earned in small units that need to be accumulated through continued performance of the behaviour to earn the outcome. Thus, there is no need to give a tangible reward every time the target behaviour occurs. The token is sufficient. Third, while a reinforced behaviour occurs with high frequency, initially reinforcing every time with a tangible reward may be impractical and may cause satiety effects, where the reinforcer starts to become devalued and loses its ability to strengthen the behaviour. However, tokens, like money, seem resistant to satiety effects and so can be used much more freely. Finally, token systems offer the opportunity to remove, rather than just award them. Such response costs can be considered more acceptable and ethical than completely withholding a material reinforcer or privilege. The individual simply has lost some of their accumulated tokens and so has to resume the desired behaviour to re-earn them. Slide 17 To work effectively, a token system needs ﬁve basic components. First, the nature and value of the token must be explicit and understood. Second, an accurate and transparent means of recording tokens earned, spent, or removed. Transcripts by 3Playmedia Week © King’s College London 2018 7. Third, there must be clear and valued actual reinforces that can be acquired through accumulated tokens. Fourth, the rules governing the earning of tokens and possibly losing them must be clear. And ﬁfth, consistent implementation, particularly not allowing access to the tangible reenforcer other than through the tokens. Slide 18 Token economy systems in mental health gained in popularity from the 1970s onward and continued to be used, more or less widely, in a range of inpatient and outpatient settings, typically as part of a wider therapeutic regime. They are also often used in the management of children and adults with learning disabilities and neurodevelopmental disorders, as well as in the areas of addiction or substance misuse, to develop and support abstinence. Although evidence from single cases, case series, and small clinical trials abound, evidence of large-scale, well-conducted clinical trials is relatively rare. Generalising across disorders and settings is therefore not possible. A recent review by Faith Dickinson and colleagues in 2005 reviewed published studies on the use of token economies in long stay psychiatric hospitals. Most of these studies were carried out between the 1960s and 1970s, with only a few later ones. A total of 13 control trials were reviewed and found that the token systems were generally successful in increasing adaptive behaviour, particularly around self-care and other activities of daily living and engagement with activities. Let’s look at a recent published study, this time in the context of short-term or acute hospital admission. Slide 19 Assaults by patients on other patients and staff can be a signiﬁcant problem in acute hospitals that admit extremely ill and distressed people with severe mental health problems. In the past, such patients would have been managed with tranquillising medication to reduce such behaviour. This study, within a single inpatient unit, compared the records for assaults for a period of 12 months before a token system was put in place and for two years afterwards. The speciﬁc system was designed on an individual basis within the overall ward environment. In other words, each person had their own identiﬁed behaviours and the rewards that they could choose to work for. A token system comprised ink stamps on a record sheet or poker chips. Tokens could be traded for leave passes, extra smoking breaks, or trips to the cinema, or a range of items in the shop, such as cards or snacks. Tokens were awarded for behaviours that were deemed therapeutically useful, including being on time for ward activities, improved self-care, or involvement in social activities. In this particular token system, a loss of tokens was used for violations of safety rules, such as smoking in the room or violence against another person or property. This led to the loss of all accumulated tokens. The results are summarised in the graph. This shows a reduction in total injuries following the introduction and use of the token system, even though it was not aimed speciﬁcally at aggressive behaviour. The reduction was a substantial 33%, compared to the time before the token system was used. The largest change was in the injuries between patients that fell by 48%. Injuries sustained by staff were relatively rare and showed only a small drop. However, the severity of the injuries was reduced. The average number of staff hours lost due to injury fell from just under three hours per week on average to less than 10 minutes, a drop of 93%. Transcripts by 3Playmedia Week © King’s College London 2018 8. Although showing some useful improvements in reducing challenging behaviour, this study still provides only weak evidence, typical of many trials of token systems. It used a simple AB design, which cannot control for other factors that might have changed over the study period, such as change in staff or other adjustments that may have affected patient behaviour unrelated to the contingency management approach. Larger, better controlled and multi-center trials are typically needed to provide the best evidence of effectiveness of a complex intervention in a natural setting. Unfortunately, such trials remain rare. Transcripts by 3Playmedia Week © King’s College London 2018 9. Module: Psychological Foundations of Mental Health Week 5 Psychological therapies: from behaviour modiﬁcation to behaviour therapy Topic 1 The ﬁrst wave - behavioural psychotherapy - Part 2 of 3 Professor Richard Brown Department of Psychology, King’s College London Lecture transcript Slide 3 Next, let’s turn to the so-called British tradition of behavioural psychotherapy. This developed as a direct descendant of the work of an early American psychologist, Mary Cover-Jones. Her work, in turn, was based on that of Watson and Pavlov, rather than Skinner. Another important inﬂuence was the neobehaviorist Clark Hull, who emphasised the role of motivation or drive in shaping behaviour. The concept of reinforcement and operant methods were less central to what became the British tradition. Rather, it drew more on the principles of stimulus-stimulus learning, using deconditioning and extinction as a therapeutic tool. Unlike in the US, the target of the treatment was less on changing behaviour than in addressing emotion, in particular, emotional distress, fear, anxiety, and worry, so-called neuroses. Hence the term was called behaviour therapy, rather than behaviour modiﬁcation. A major driver for this development with the return of thousands of soldiers from the First World War suffering with so-called “shell shock” or “war neurosis.” In the UK alone, 80,000 soldiers were identiﬁed as shell shocked by the end of the war, probably only a fraction of the number left with signiﬁcant psychological trauma. Treatment, when offered during the war itself, was designed to return the soldiers to ﬁghting as soon as possible. After the war ended, the challenge became ﬁnding a way to offer care and treatment for large numbers with lasting psychological problems. The Maudsley Hospital here in South London, next to the IoPPN was established in 1907 as the ﬁrst modern hospital for the treatment of mental illness. By the time it was built and opened in 1916, the First World War was well underway, and its ﬁrst role was in the assessment and treatment of soldiers suffering with shell shock. We see here a couple of photos taken at the time, the one on the right showing soldiers engaged in a workshop doing occupational therapy, one of the standard treatments of the time the psychiatric disorders before modern behavioural approaches were developed. It is worth noting that the Maudsley Hospital also had the ﬁrst psychiatry medical school in the UK. It trained doctors and nurses, but also carried out research into the causes and treatment of mental illness, including shell shock. Transcripts by 3Playmedia Week 5 © King’s College London 2017 1. Fully established in 1924, that medical school eventually became the Institute of Psychiatry, Psychology, and Neuroscience. Today, we carry out the same roles as the original medical school and continue to work in close partnership with the Maudsley Hospital with which it shares a location in South London. On the KEATS pages, you will ﬁnd an interesting leaﬂet produced from a photographic exhibition in 2015, illustrating the Maudsley at war, the story of the hospital during the Great War. Slide 4 In week 1, we learned about Pavlov’s pioneering research in the classical conditioning in dogs. That evidence was taken up by the behaviourist J.B. Watson as a possible mechanism for the acquisition of neuroses and phobias. We learned how he tested this hypothesis in the famous or infamous case study with the baby Little Albert. You had a chance to review the evidence and decide for yourself how convincing this was as a demonstration of a classically conditioned fear response and its generalization to other objects. However, the study was highly inﬂuential as evidence to support a behavioural learning model of fear, without the need to infer an inner or subconscious process ask Freud did. It also has stimulated new approaches to the treatment of neurosis. Before we come on into the early clinical developments from Watson’s work, we should consider an important question about Little Albert. What happened to him? Or more precisely, what happened to any conditioned fear that he had acquired over the days of the study? This is an important question, not just out of concern for the child, but for our understanding of how fear and fear behaviour develops into a problem that can remain with a person for their whole life. We will remember that experiments in classical conditioning in animals show that the conditioned response does not usually persist forever but tends to extinguish when a conditioned stimulus ceases to be paired with the unconditioned stimulus. Given that Watson and Rayner stopped pairing a rat with a loud noise, we would expect that any conditioned fear that Albert had acquired would soon go away. We will never know for sure whether this was what actually happened. Years later, attempts to identify and follow up Little Albert as an adult have not been wholly successful. The most likely candidate was Douglas Merritte, who sadly died aged 6 of a neurological condition. Incidentally, this makes it hard to know whether Watson and Rayner’s experiment tells us anything about fear conditioning in normally developing babies. However, if his fear did persist, despite expectation that it would extinguish, we would need to ﬁnd a way to explain it. This would not mean that classical conditioning was unimportant in the acquisition of the fear but that other processes may be necessary to explain its persistence. Conversely, if the fear did not persist, it would suggest other factors must be at play to explain the developing of lasting neuroses and phobias. We will turn to the distinction between onset or etiological factors and maintaining factors later on. They have an important place in modern cognitive theories of mental health and particularly in the identiﬁcation of targets for treatment. Slide 5 You will recall that Watson and Rayner explained how they had planned to reverse or decondition Little Albert’s fear response. However, before they could do this, he was unexpectedly discharged from the hospital. Not only did this leave Little Albert at risk of a lasting distress, but it prevented Watson from further testing his theory, that not only could a neurosis be explained using behavioural theories and created using behavioural methods, but that the same theories and methods could quickly and easily remove a learned neurosis. It would seem that Watson never went on to test this latter prediction himself, or if he did, he never wrote about it. However, his work directly inspired others to explore the potential of behavioural approaches as treatments. One of the earliest pioneers was the American psychologist Mary Transcripts by 3Playmedia Week 5 © King’s College London 2017 2. Cover-Jones, who had heard Watson present his work and theories. In the course of her research in what would now be called behavioural genetics, Cover-Jones investigate the potential of classical conditioning to remove fear or phobic responses in young children. In a famous report in 1924, she describes in detail one of a series of such experimental treatments. The subject was a young boy, just under three years old, called Peter. Otherwise healthy, he showed considerable fear when presented with white rabbits, white rats, fur coats, and rugs, but not to other objects, such as wood rabbits or other toys. This similarity to Little Albert was striking to Cover-Jones and suggested that the rabbit and other furry items were a conditioned stimulae. As there was nothing in his history to say that he had come across rabbits or rats before, Cover-Jones assumed they were what she called a “transferred fear,” or what we would call generalised conditioned response. Cover-Jones worked with Peter over a period of months, sometimes twice a day and sometimes once, often with large gaps in between. She was careful to call this study an experiment, not a treatment. Over this time, she deﬁned a series of stages to chart Peter’s growing ability to tolerate the presence of the rabbit and be less afraid. These varied from step A, where the rabbit anywhere in the room caused a fear reaction, to step Q, where Peter allowed the rabbit to nibble his ﬁngers. Slide 6 Cover-Jones reported Peter’s progress with the graph shown here, with the degree of toleration on the vertical y-axis and the sessions on the x-axis. Note, the period of time between sessions was not constant and included long gaps of up to two months. By using the term “degree of toleration,” Cover-Jones was keeping with her behavioural background, measuring what could be observed-- Peter’s ability to tolerate the rabbit’s presence, rather than his emotional state. Cover-Jones showed her careful approach by ﬁrst checking that this toleration series did actually amount to a systematically ordered stage of change by getting six other people to rank the levels in order. This showed a high level of agreement between the different raters. Today, we would call a tolerant series an exposure or stimulus hierarchy. She started by presenting the rabbit to Peter in an unthreatening way, something that we would now call exposure, allowing them to get used to it over a period of time. This also included allowing Peter to see other children playing with the rabbit without fear, an approach that is still used and called modelling, a form of observational or social learning. This stage of the experiment showed that Peter progressed relatively quickly, to the point where he was able to touch the rabbit. This period is shown in red on the graph. Peter then became ill and did not return for two months, at which point his fear behaviour had returned to its original level, and he was unable to tolerate the presence of the rabbit. Rather than an extinction process, this was attributed to a recent experience where Peter was badly frightened by a large dog. Slide 7 Cover-Jones then moved to a different approach based directly on classical conditioning. Her aim was not to change the nature of the conditioned stimulus-- the rabbit-- but the nature of the conditioned response to it, from a fearful one to a positive or at least a neutral one. She ﬁrst needed an unconditioned stimulus that was associated with a positive unconditioned response. Cover-Jones chose sweet food, biscuits and candy. After establishing the Peter liked sweets, Jones set about preparing the presentation of a rabbit with the presentation of the food. However, rather than just throwing the rabbit in Peter’s lap with a biscuit, as Watson might have done, Cover-Jones presented the rabbit in a cage some feet Transcripts by 3Playmedia Week 5 © King’s College London 2017 3. way from Peter when giving him the sweet. The distance was as close as they could get without provoking a response in Peter. This was repeated over several occasions, gradually bringing the rabbit closer, with corresponding reduction in Peter’s apparent fear. Thus, the unconditioned response to the rabbit changed to the point where Peter allowed the rabbit to play with him in his playpen. The study improvement is shown in the blue area during the period of 25 sessions. Following a temporary reversal when Peter was bitten by the rabbit, he continued to progress to show even greater tolerance, allowing the rabbit to nibble his ﬁngers without apparent fear. A ﬁnal research innovation that Cover-Jones used in this case study was the use of physiological methods to measure fear. In addition to measuring Peter’s behaviour, she also measured his blood pressure as an objective measure of his physiological arousal or visceral response. She showed how it became elevated when he showed signs of fear, although she did not go on to use it as an objective measure of change over time with the experimental treatment. Slide 8 Although an uncontrolled single case, this report established many of the principles and methods that would inﬂuence behavioural psychotherapy in the decades to come, many of which are still used today. Let’s summarise Mary Cover-Jones’s legacy to these later developments. The ﬁrst is that she took theory and empirical evidence and applied it to the treatment of a psychological problem, in this case, irrational fear. This is a very different approach to what went before in contemporary psychoanalytic methods of Jung and Freud. Although these also had an underpinning model or theory, they were not empirically based. The contrast between the two reﬂects the difference between the empiricists and the rationalist traditions that have run through science since its outset. Cover-Jones’s experimental approach precedes today’s scientiﬁc approach to understanding the processes underpinning mental health problems and the development and evaluation of new therapies. In terms of methodology, in the single example of Peter, she gave us a number of approaches that continue to be used today. She took existing ideas and demonstrated the potential of deconditioning, pairing a feared stimulus with an alternative non-feared one. She used the principle of exposure, she used a graded approach to exposure, what she’d call degree of tolerance that today we call stimulus hierarchies. She used observational techniques, including others showing non-fearful reactions to the feared objects, today called modelling. In terms of assessment, she used proven, reliable methods to deﬁne and measure change. And ﬁnally, she introduced the possibility of using physical measures as objective non-behavioral indicators of fear, what we now call psychophysiology. Slide 9 Although widely recognised today, this groundbreaking study was largely forgotten for many years. This may be because Cover-Jones was a woman at a time when science was still dominated by men or because operant conditioning replaced classical conditioning as the dominant model within US behaviourism. Nevertheless, her contribution was eventually recognised in later years, and she has been called the mother of behavioural psychotherapy. She died in 1987 at the age of 91 after a long career in science. The majority of this work was based on the Oakland growth study, a longitudinal investigation of 200 children as they grew up to be adults, measuring their physical and psychological health and its impact on later problems. In an article that she wrote in 1976, she reﬂected on how this longitudinal approach inﬂuenced her thinking on the historical reports on Peter and on Watson’s Little Albert. We see here an extract from that paper. Transcripts by 3Playmedia Week 5 © King’s College London 2017 4. Slide 10 Following Cover-Jones’s work, one of the pioneers of the 20th century behaviour therapy was the South African psychologist Joseph Wolpe. Most of his early work was in animals, developing experimental models of neuroses by conditioning fear, and then ﬁnding ways to reduce or remove it. He use punishment techniques that would not be permitted today. However, he is most justiﬁably famous for how he applied this early work and other behavioural research to the treatment of neuroses. His work was based on the concept of what he called reciprocal inhibition. This proposed that it was impossible to be both anxious or afraid, while at the same time carrying out behaviour that indicated the opposite, in other words, relaxation and fearlessness. Wolpe started out using assertiveness training as a way to address problems with shell shocked soldiers that we would now see as social anxiety. In such patients, developing positive and assertive social interaction was seen as incompatible to the fearful withdrawal in many life situations that these ex-soldiers showed. However, his clinical studies quickly suggested that a more effective method could be found in those ﬁrst described by Mary Cover-Jones with her case study of Peter and the rabbit. Slide 11 Many of the approaches developed by Wolpe owe their origins to Cover-Jones’s writing, combined with Wolpe’s own pioneering work in animals. Techniques that Wolpe affected included the identiﬁcation of clear stimulus hierarchies. This was a graded list generated by the patient and the therapist ahead of treatment. It listed situations from the easiest to the most difficult, with a series of steps in between. Here’s an example of a hierarchy for someone with a fear of spiders. These were used as the basis to systematically expose the patient to a situation, starting with the easiest and waiting for any initial levels of anxiety to subside. Critically, the approach required that the patient did not engage in any behaviours that would reduce the fear, such as backing away, closing their eyes, trying to distract themselves, and so on. Inhibition of such behaviours was called response prevention. The idea was that the patient experienced the anxiety and learned that it would reduce on its own. To facilitate this process, Wolpe developed a simple numerical scale that patients could use to rate their own levels of anxiety, so-called Subjective Units of Distress or Discomfort Scale. Because there was no behaviour to observe during exposure, this subjective rating became the only way for the patient and the therapist to monitor change. The scale typically has 100 points, from 0, no distress, to 100, the worst possible stress, with labels at intermediate points on the scale, typically in steps of 10. Sometimes the scale goes from 1 to 10 and supporting images are used to make them even easier to use, especially for children like the one shown here. Once the patient was comfortable with low distress in the easiest situation on the hierarchy, the treatment moved on to the next step and so on. Depending on the nature and severity of the phobia, this would be done in a single session or extended over several sessions. Prior to exposure treatment, patients were typically taught relaxation. This provided them with the skill to reduce their level of anxiety, both subjectively and also the physical symptoms, before enduring exposure. This builds on the idea of reciprocal inhibition. Being relaxed is obviously incompatible with being afraid. Being relaxed in the presence of a feared object or situation was felt to enable counter-conditioning. Wolpe termed this overall treatment approach “systematic desensitisation.” Transcripts by 3Playmedia Week 5 © King’s College London 2017 5. Slide 12 Systematic desensitisation became a mainstay of behavioural therapy for the treatment of a range of anxiety disorders, from phobias to social anxiety to obsessive compulsive disorders. Over time, the basic techniques became a further reﬁned and adapted to increase its effectiveness, but also to allow it to be applied to a wider range of problems. Even early on in its development, Wolpe introduced the approach of imaginal desensitisation. It is the nature of anxiety that the stress can occur when a person imagines a step on the hierarchy, as well as when it is actually experienced. This is further evidence from brain imaging studies that show the physiological equivalence of real and imagined exposure to fearful stimulae. This is a clear example of the way in which our internal state, our memories, perceptions, and expectations, can inﬂuence our emotions based on a common neuronal substrate. One implication of the power of the imagination is that therapy can be done by asking the person to imagine exposure, rather than actually happen in real life or in vivo. This has a number of advantages. First, it broadens the range of target problems that can be treated. A person who is afraid of heights does not necessarily have to be taken up tall buildings. Second, it allows the patient to practise exposure more easily, including between therapy sessions. Imaginal exposure is sometimes called in vitro, implying that the stimulus is not actually present. Imaginal desensitisation became an important therapeutic approach, although subsequent evidence suggested that it was less effective than in vivo exposure when the latter was practical. A disadvantage of the imaginal approach, however, is that it only really worked in patients who were able to conjure strong and convincing visual images, something that not all people can do. Increasingly today, virtual reality techniques are being used to bridge the real and the imagined, opening up new opportunities to reduce anxiety. Today, classic systematic desensitisation approaches are rare as the main component of treatment, having been replaced by a wider a more effective cognitive and cognitive behavioural approach. However, some still incorporate elements of Wolpe’s original methods. We’ll look at an example shortly. Slide 13 First though, let’s look at another approach ﬁrst introduced by Mary Cover-Jones-- deconditioning by observation. In other words, that it is possible to reduce our responses to fearful or anxiety provoking situations by watching another person experience it, rather than experience directly or even imagining it. The ability to learn by observation formed the foundation in what came to be called social learning theory, and its pioneer was the Canadian psychologist Albert Bandura. His work formed another important strand in the so-called cognitive revolution in the 1960s that we looked at in week 1. Simply stated, social learning theory proposed that we can and do learn important behaviour from watching other people and observing the consequences of their actions. In other words, we can learn without being directly reinforced ourselves something that ran contrary to operant models at the time. Social learning theory was an important inﬂuence on later developments in the area of social cognition and even mirror neuron systems that you learned about in previous weeks. We also saw in week 1 the powerful example of observational learning in the problem solving ability of Goffin’s cockatoo’s. One of Bandura’s most inﬂuential series of studies was the Bobo doll experiment of the early 1960s in which he investigated the inﬂuence of observation on the development of aggressive behaviour in young children. A Bobo doll is a large inﬂatable doll, weighted at the bottom, that stays upright when played with. In the classic study from 1961, a group of 72 girls and boys, aged 3 and 1/2 to just under six years, were randomly assigned to groups with equal numbers of boys and girls in each. Before playing Transcripts by 3Playmedia Week 5 © King’s College London 2017 6. with the doll themselves, 2/3 of the children ﬁrst watched an adult, either a male or female, playing with the doll. For half, the adult played aggressively with the doll, such as hitting it with a hammer. For the others, the adult played nonaggressively. For the remaining third of the children, they did not see an adult playing with the toy. There were other parts to the study and its design. However, the main ﬁnding to note here is that the children who observed aggressive play towards the doll were more likely to imitate the aggression when they came to play with the doll themselves. There is a link to a short video showing Bandura talking about his experiments on the KEATS page. This study, and others like it, have had a strong inﬂuence on the debate around exposing children to violent behaviour on TV, and more recently, in video games. Slide 14 Bandura’s experiments were about the transmission of violence. However, fear behaviour has a social function. We communicate danger to others so that they can take appropriate action. Because of this, it is logical that we can learn to be frightened by observing others expressing fear in a speciﬁc situation. Indeed, this is one of the possible mechanisms for the intragenerational transmission of fear. A child who observed his or her own parent demonstrating a fear of spiders is more likely to develop such a fear themselves. You learned in week 3 about the role that observation and instruction and can play in fear acquisition. I remember an incident with my own daughter when she was about 8. We were on holiday by the seaside, and she was with a group of children, jumping from a high rock into a deep pool of water. Although scary, it was also exciting, and my daughter jumped in with screams of mixed fear and delight. She then made friends with another girl of her own age. That girl refused to jump, showing evident fear and voicing her distress. My daughter then refused to jump anymore. To this day, almost 20 years later, she will not jump into deep water, other than in the swimming pool. However, just as observational learning has the potential to reinforce fearful avoidant behaviour and aggression, it can also do the opposite. In therapeutic modelling, the patient observes another person in a situation that they themselves ﬁnd distressing, just as Cover-Jones did in some of her sessions with Peter. Later research has shown that the effect of modelling is even more powerful when the other person was themselves afraid of the same thing. This was termed “coping modelling,” as compared to the situation where the other person had no fear, so-called “mastery modelling.” Seeing someone doing something despite being afraid is a more powerful way to help us overcome our own fears than seeing someone fearless doing the same thing. Transcripts by 3Playmedia Week 5 © King’s College London 2017 7. Module: Psychological Foundations of Mental Health Week 5 Psychological therapies: from behaviour modification to behaviour therapy Topic 1 The first wave - behavioural psychotherapy - Part 3 of 3 Professor Richard Brown Department of Psychology, King’s College London Lecture transcript Slide 3 Having covered some of the basics of behavioural psychotherapy approaches, let’s look at a couple of examples of their application. First, the treatment of dental fear or, in this case, fear of injections as part of dental treatment. Dental injection fear and fear of the dentist more generally is one of the most common fears. A study of almost 2,000 Dutch adults, aged between 18 and 93 years, found that dental fear was the fourth most common fear, after fear of snakes, heights, and physical injury. Almost a quarter of Dutch adults reported dental fear. A fear becomes a phobia when the fear has a more significant impact on the person and their life, such as avoiding the dentist and suffering poor dental health and even pain as a result. When considered at this level, dental phobia was the most common reported, with 3.7% meeting criteria for diagnosis. Slide 4 Here is a simple model. Dental fear, when present, can stop people attending the dentist for treatment. When they stop attending, dental problems can build up. Eventually, treatment becomes inevitable. Because of the need for more invasive treatment, the experience is more distressing, reinforcing the fear and making it even harder the next time. This is an example of a vicious cycle, which can maintain fear. It can develop into a major problem, adversely affecting the person’s health. The behavioural response to the fear, delayed visiting, is a form of avoidance. For some people, the avoidance takes the form of never visiting the dentist, rather than just delaying. Because of this problem, dental phobia and the fear of drilling and dental injections has long been a target of behaviourist approaches, such as systematic desensitisation, with dentists themselves trained to deliver the interventions to remove obstacles to treatment. Slide 5 Let’s look briefly at a trial that used systematic desensitisation, along with other techniques, to reduce dental injection fear and improve the chance of successful dental treatment. This trial by Transcripts by 3Playmedia Week © King’s College London 2018 1. Lisa Heaton and colleagues recruited people who were identified as having a significant fear of dental injections, leading to avoidance of required dental care. Rather than seeing a psychologist, the intervention was delivered primarily via computer, an increasingly used technique that offers many potential economic and practical advantages. The programme followed a standardised, rather than individual, stimulus hierarchy, shown here. The stimuli were videos of a person going through various preliminary stages leading up to the moment before an actual injection. As in the original methods of Wolpe, the research participants were first trained to relax and encouraged to use relaxation as they worked through the video. They could do this at their own pace over about a 30 minute period, recording their fear levels or distress as they did so at each step. The computer only allowed them to move on if they reported their anxiety to be low enough. On- screen feedback was given, encouraging the person to move to the next step or to go back one, if they had moved too quickly through the hierarchy. Slide 6 The figure on the left shows what’s called a consort diagram for the study. This is a standard way of reporting the flow of participants before the trial, as they were first screened, to make sure they met the criteria and after they’d been randomly allocated to receive the computerised treatment or a leaflet condition. We see that over 250 people were initially screened and 84 eventually entered the trial. The rest either didn’t meet the criteria for the trial or refused to take part. Participants were randomly allocated to receive either the active treatment or a simple leaflet condition containing information about the procedures of how to manage pain. Most of the participants completed the treatment and were followed up to see what changes could be measured. Three different self-report measures of dental anxiety were used. In the figure on the right, the white bars show what happened to the average response of those who were provided with the information leaflet, the control group. This showed a small but not significant reduction in dental fear on each of the measures. As expected, those who received the desensitisation treatment, shown by the black bars, showed a greater average reduction in self- reported fear. Such self-report, however, is not the same as actual behaviour, in this case, reduction and avoidance. In this respect, the programme also did better. 34% of those who had had the programme went on to have a dental injection and treatment, compared to 17% of those who had had the leaflet. However, although encouraging, this difference was not statistically significant, due to the small sample size. One point to note about this study was that the intervention was not purely behavioural. As well as the relaxation and graded exposure elements, the video also contained information on coping strategies to be used before and during the procedure. This means that the intervention is better considered as a combined behavioural and cognitive one. This pragmatic approach is fairly typical of modern behavioural interventions. They continue with some of the tried and tested principles of behaviour therapy, but supplement them with additional non-behavioral features that add to the overall clinical value. We will come back to some of the cognitive approaches in the next topic. Transcripts by 3Playmedia Week © King’s College London 2018 2. Slide 7 Here’s one more example of a desensitisation approach from a single case study of a patient with cockroach phobia, carried out in Spain. The figure shows the change in the individual patient’s ratings of fear and avoidance over the treatment period and after treatment had ended. The study was also an early attempt to explore potential use of mobile technology and virtual reality to treat phobias. In this case, treatment involved the patient interacting with a mobile phone game involving animated cockroaches appearing on the phone screen. We see on the left the initial baseline fear of the patient to cockroaches, over the 14 days before treatment started. She then spent about one week playing the mobile game on a daily basis. We see that her fear ratings dropped considerably, from 10 out of 10 at the start of the week to six by the end. Further improvement was noted after the second week. This was not only maintained when the patient was followed up later, but continued to improve over the coming year, until fear dropped to zero. We also see in the graph below the ratings of avoidance tendency reported by the patient across the same period. We see a similar pattern of improvement, except this avoidance tendency was eliminated completely by the second session. This shows a common finding in many psychological treatments, namely that the rate of change of different outcome indicators can vary, sometimes called response desynchrony. Another feature of this treatment involved the use of augmented reality between the first and the second week. This involved the patient viewing a live video of their hands, over which a virtual cockroach walked. The rating before and after this exposure session are shown highlighted with red circles. This produced an additional drop in fear across a single session and a considerable drop in avoidance. Slide 8 Before we end this topic, we are going to expand on something that we touched on in the examples Ann mentioned previously, namely avoidance behaviour. We will look at how avoidance learning was initially explained by behavioural theories, but then how increasingly, cognitive processes were required to match the evidence. First, though, let’s talk about a related behavioural concept, that of escape. In behavioural terms, escape is a response that distances us from an ongoing unpleasant or aversive event. When confronted with a dangerous situation, we typically experience fear and respond by taking ourselves from the danger as quickly as possible. This is a completely normal response and one shaped by evolution in all animals, many of which have their own species-specific reactions. For early behaviourists, of course, the emotional experience would have been less important than the observable behavioural response. Whether animals experience fear as we do may be questioned, but they typically show the same physiological reactions and behave in a way that is consistent with fear. Here’s a simple example of escape, showing an escape response in ABC terms. The antecedent is the aversive stimulus, in this case, a snake that suddenly strikes out as we’re walking through some long grass. The behaviour is running away. And the consequence is survival. For humans, escaping from real danger, whether a flooded river or wild animals, has obvious survival value. However, survival is enhanced further if we learn how to avoid the danger in the first place, rather than escape whenever encountered. In this case, the behaviour, avoidance, can be seen as a response to a situation associated with danger, rather than actual danger. Transcripts by 3Playmedia Week © King’s College London 2018 3. For example, if we’ve learned that snakes are most likely to be found in long grass, then long grass can become the antecedent, in ABC terms. By avoiding the long grass, we continue to survive. We will never know whether there was a snake present, but this doesn’t usually matter. In behavioural terms, avoidance can be seen as either active or passive. If we find ourselves in long grass and leave straight away, this is active avoidance. We are doing something that prevents an aversive event that would happen if we did nothing. By leaving the grass, we avoid the encounter with a snake. In contrast, in passive avoidance, we are avoiding the situation in which an encounter with a snake could happen. In other words, we see the long grass, and we do not enter it. So while escape responses distance us from ongoing aversive events, avoidance, both active and passive, results in the omission of a future aversive event. Both escape and avoidance are central to our understanding of aspects of mental health. Behaviour that reduces the risk of being bitten by a snake is obviously useful or adaptive if we live in a part of the world full of dangerous snakes lurking in the grass. However, it is less helpful in the UK and other parts of the world, where snakes and similar natural dangers are rare. Here, avoidance of long grass will be seen as an example of maladaptive behaviour, in other words, a behaviour that, in avoiding a very unlikely future averse event, causes the person to adjust their actions and limit their opportunities. Maladaptive avoidance behaviour plays a central role in anxiety disorders, but also makes a contribution in other disorders, including depression Slide 9 How do we learn to avoid dangerous or potentially dangerous situations? Can we explain avoidance learning in simple classical or operant conditioning terms, in other words, by stimulus-stimulus association learning or processes based on reinforcement? The first theories of avoidance learning were based on Pavlovian classical conditioning. In these explanations, we learn to associate a real aversive, unconditioned stimulus, the snake, in our previous example, with a previously neutral, unconditioned stimulus, grass. The unconditioned response is avoidance. By this pairing, the grass is thought to become an aversive Condition Stimulus or CS negative, leading to the conditioned response. In a laboratory setting, a dog naturally lifts its leg if given a mild electric shock to its foot. The shock is the aversive unconditioned stimulus and the leg lift the unconditioned response. If a bell is paired with the shock, the bell can become an aversive, conditioned stimulus, and the leg lift becomes a conditioned response. Although this is seen in practice, it is unclear whether the mechanisms are the same as other learning through classical conditioning. Critically, in conditioning a salivation response to a bell, the dog gets the food regardless of whether they respond. The outcome remains the same. With a conditioned avoidance response, however, the exposure to the aversive outcome depends on whether or not the animal makes the defined response. This situation of response contingent outcomes seems to be more like an operant or instrumental conditioning, as we see here with a simple, three-part contingency. The antecedent is the bell. The response is the foot lift. And the outcome is no shock. However, this poses the question, what is the reinforcer? For Watson and the early behaviourists, the reinforcer needed to be observable, just as much as the stimulus and the behaviour. In avoidance learning, however, the potential reinforcer is the omission of a pending aversive event, something that has not yet occurred. Transcripts by 3Playmedia Week © King’s College London 2018 4. This required a reconsideration of the nature of reinforcers in avoidance learning. While as humans, we can see the avoidance of future negative outcomes as reinforcing, a seeming ability to predict the future was not compatible with early behavioural models of the time. Some other, more acceptable intervening variable was required. The variable that was proposed was fear and led to what was termed two-process theory of avoidance learning. Slide 10 The two-process theory of avoidance learning is most closely attributed to the American psychologist Orval Mowrer and first published in 1947. While fear, as an internal and unobservable emotional state, was not assumed to be a necessary construct for early behaviourists, it provided a means to explain laboratory studies in animals and our human experience. It could be included in a behaviourist framework, if the emotional state was assumed to be a covert behaviour and a component of the total behavioural response, both observable and unobservable. As the name of the model suggests, two-learning processes were proposed to explain avoidance learning. The first was a stage of Pavlovian stimulus-stimulus association, the pairing of neutral and aversive stimuli with an aversive outcome. In this way, the previously neutral stimulus acquired aversive properties. Through such pairings, a natural emotional response, fear, is also conditioned, not just a physical response. In the second stage, operant processes take over to reinforce the behaviour. The reinforcer is reduction of fear itself, not by the non-delivery of the aversive outcome. Thus, an increase in avoidance behaviour to the antecedent, the conditioned fear stimulus, is negatively reinforced by the reduction in fear. Slide 11 This model was hugely influential, not just in terms of learning theory, but also how it impacted on behavioural approaches to treating anxiety. As with any good theory, it not only seemed to provide a satisfactory explanation for existing evidence, but also predicted novel findings. Again, however, as with any theory, evidence accumulated that it found difficult to explain. We don’t have time to go into these in detail, so we will just summarise them here. You will find more information, along with a fuller discussion of avoidance learning, in one of this week’s key readings by Angelos Krypotos and colleagues. In brief, here are a few of the most common criticisms. These do not say that the model is wrong and does not apply in some situations, only that it is not a full explanation, specifically, that the two processes are not always necessary to explain avoidance learning and the maintenance of avoidance behaviour. The first criticism is that fear does not seem to be necessary for avoidance behaviour to continue, once established. If Pavlovian methods are used to decondition the fear response, the avoidance response does not necessarily stop as a result. For example, a dog that has learned to jump a barrier to avoid shock when they heard a buzzer will continue to jump, even after the shock apparatus is turned off. Critically, the animal no longer shows a fear response to the previous aversive conditioned stimulus, either in terms of behaviour or physiological reaction. Despite this, it will still act to avoid what is now a seemingly neutral stimulus. In other words, avoidance responses can become habitual and no longer contingent on reinforcement for their maintenance. A second criticism came from clinical studies based on applying the two-process model. The model suggested that desensitising fear by exposure would only be effective if terminating the aversive stimulus occurred when fear or distress had reduced to near zero. If exposure was terminated Transcripts by 3Playmedia Week © King’s College London 2018 5. early, while the person was still experiencing distress, the fear and urge to avoid would be reinforced. In practise, however, research carried out here at the IoPPN by Jack Rachman and colleagues in the 1980s show that exposure was equally effective, regardless of whether it was terminated when distress levels were high or low. Third, animals can learn to avoid a negative outcome even without an aversive stimulus. For example, when rats are given a shock at a fixed time interval, they learn to avoid the shock by moving into a different part of the cage. The absence of the conditioned stimulus makes the explanation of a first stage Pavlovian process problematic. Similarly, in human avoidance behaviour, there is often no evidence of an initial conditioning stage. For life or death situations, the need to rely on Pavlovian processes for learning would seem to make little evolutionary sense. Even in non-life threatening situations, such as a fear of spiders or dogs, people can often not recall an early episode of ever being badly frightened, marking the start of a process of avoidance learning. Other process are often suggested, such as observational learning, seeing someone else to be scared, or instructional learning, being told that dogs are scary and to avoid them. These are both examples of social learning that lies outside of the two-factor model. Slide 12 This brings us to the more cognitive explanations of avoidance behaviour, both in animals and humans. These focus on the role of information about the aversive stimulus or outcome, rather than the events themselves. Again, we will look at just a couple of examples here, with a fuller description available in your reading for the week, in the paper on avoidance learning. The first piece of evidence comes from highly influential studies in the 1970s by two American psychologists, Robert Rescorla and Allan Wagner. They showed when a CS was inevitably followed by an aversive US, there was actually less learning, that is avoidance to the CS, than when there was an element of what they called surprise. In other words, the US sometimes occurred in the absence of the CS, and sometimes, the CS was not followed by the US. This created an element of uncertainty or a prediction error in information about the probability of an aversive outcome. When there was no uncertainty, prediction error, there was reduced or no learning. It seems that such uncertainty is itself aversive. In human terms, people waiting for some possible bad news often report that the period of waiting and not knowing is worse than the eventual outcome, even if that outcome is a negative one. While we might propose a range of explanations for this, research suggests that even rats prefer to know that something bad is going to happen, rather than wait without being sure. For example, in an experiment, rats were given a choice of staying in one of two compartments. In one, a light came on to signal a shock, from which it could not escape. In the second, the rat was shocked equally often, but without a warning. The rats chose to spend time in the box with the light. The explanation comes from the informational value of the light or, more precisely, the informational value of the absence of the light. This absence of light could be considered a safety signal, reducing uncertainty and therefore fear. Slide 13 An extension of the idea of safety signals is that of safety behaviour, a concept that has become an influential one in understanding the maintenance of maladaptive avoidance or resistance to treatment using exposure therapy. Let’s go back to our snake in the grass example. Sometimes, we can’t avoid walking in the grass, even if we have been scared before. Transcripts by 3Playmedia Week © King’s College London 2018 6. In the same way, someone afraid of flying may have to do so occasionally. In such situations, it seems sensible to take measures that help us feel safe. Obviously, when we fly, there is almost no real uncertainty. The chance of the plane crashing is so small as to be almost immeasurable. Similarly, when we walk in a London park, we are unlikely to tread on a poisonous snake. Nevertheless, mere probability is not enough to conquer fear. Remember, human beings are not rational. Because even a small chance of something happening is too much for some people, they will do things that seemingly reduce the uncertainty even more, with the aim of helping them feel safe. Walking through the grass, we may look carefully before we take every single step, so-called hypervigilance. We might make a noise to scare away the non-existent snake or carry a stick to beat it off in the very unlikely event that we see one. Slide 14 Let’s look in a bit more detail at what is happening here. We are exposing ourselves to a situation that we find frightening, and, of course, nothing happens. The safety behaviour has, in our minds, reduced the uncertainty of the null accounts of an aversive event. The result is a temporary reduction in anxiety, but at a cost. We remain fearful and avoidant in the long term, relying on safety behaviours as a way to survive. Many superstitious behaviours can also be considered safety behaviours. For example, we wear a lucky charm to stop us being struck by lightning. When another day goes by without being struck, our belief in the charm is strengthened. But it does not stop us being afraid of lightning. We can become dependent on such behaviours. We come to associate them with relief from anxiety, and so they become reinforced. We learn that we are safe because of the safety behaviours. We learn that we are safe because of the safety behaviours, not because there is nothing to be afraid of in the first place. This is one mechanism that may maintain phobias. In other cases, safety behaviours can themselves become problematic and ritualised. This is what happens in obsessive compulsive behaviours, such as repeated hand-washing or checking. Slide 15 So this brings us to the end of this week’s first topic. Let’s briefly summarise what we’ve covered. First, behavioural psychotherapy arose directly from the traditions of Pavlovian and operant conditioning and their later developments. In the US, operant methods developed with their emphasis on the three-stage contingency model and reinforcement, using methods of functional analysis and contingency management to manage challenging behaviour. In the UK, Pavlovian principles were more influential, leading to the techniques of exposure hierarchies and systematic desensitisation. Extending earlier behavioural studies, the influence of observational social learning was recognised and incorporated into behaviorally based approaches. We explored the nature of avoidance learning and its role in the maintenance of a learned maladaptive behavioural response to a threat. And finally, we’ve seen an evolution of behavioural therapies and methods to include cognition, such as outcome prediction and expectancy, in driving safety behaviour, cognitions that improve our theories but also guide the development of improved psychotherapeutic approaches. This leads us into the development of cognitive therapy, the so-called second wave psychotherapies. Transcripts by 3Playmedia Week © King’s College London 2018 7. Module: Psychological Foundations of Mental Health Week 5 Psychological therapies: from behaviour modiﬁcation to behaviour therapy Topic 2 The second wave - the role of cognition and the emergence of cognitive therapy - Part 1 of 3 Professor Richard Brown Department of Psychology, King’s College London Lecture transcript Slide 3 As with any seeming new idea or approach, if we look carefully, we ﬁnd that someone else usually got there ﬁrst. When it comes to the human mind, to our thoughts, emotions, and behaviour, that person was often William Shakespeare. If there is one line of his work that captures the essence of cognitive therapy, it is this one here. “There is nothing either good or bad, but thinking makes it so.” These words were spoken by Hamlet, Prince of Denmark, when describing his country and, indeed, the whole world as a prison. In reﬂecting that it is our own thoughts that make something good or bad, he is also acknowledging that two people in the same position can have very different feelings about it. In the play, we see how this thought directly guides his emotions and behaviour. Hamlet is withdrawn and angry and probably depressed, critical of what he sees about him. What unfolds, in true Shakespearean style, is a tragedy of passion and death, and ends with Hamlet’s own suicide. Slide 4 If we were to take Hamlet’s basic premise that we can make things seem good or bad by how we think, we can propose a simple model. This starts with Hamlet’s situation, to the ultimate outcome. The intervening variable is his thoughts. These lead to the emotions and the actions. We can consider emotions as covert actions, just as Mowrer did in his two-process model. In cognitive terms, thoughts are true mediating variables. They change and transform the relationship between the input and the output, rather than serving as simple links in a chain. Shakespeare did not arrive at this model from experiments, but from his own deep understanding of the nature of the human mind and behaviour, his own and those of people around him, present and past. Of course, Shakespeare may also have been inﬂuenced by one of our friends, the Greek philosophers. As usual, they were 2,000 years ahead of us. In this case, it was not Plato or Aristotle, but the philosopher Epictetus, who died in the year AD 135. Amongst his doctrines, he proposed that Transcripts by 3Playmedia Week © King’s College London 2018 1. determining between what is good and what is not good is made by the capacity for choice, in other words, our conscious thoughts, and is not absolute. Despite this long gestation period, it was almost 500 years before Shakespeare’s insights found application in how we approach the treatment of mental health problems. This turned into the second wave of modern psychotherapies to emerge in the 20th century, cognitive therapy. As it developed, it incorporated and reﬁned behavioural approaches. And so today, the term “Cognitive Behaviour Therapy” is more typically used, or CBT, for short. Slide 5 Cognitive therapy, or CBT, is not a single entity, but describes a broad range of different therapeutic approaches. Each has been developed and applied within its own particular framework that emphasise different aspects of the mediational process and speciﬁc therapeutic techniques. Broadly, we can identify three strands that characterise the second wave approach that developed from the 1950s onwards. First, those that seek to develop a range of adaptive skills, cognitive, practical, and interpersonal. These are designed to allow a person to cope better with the situations in their life that are causing them problems. This is based on the assumption that emotional distress and negative outcomes is a result of the use of ineffective coping responses that do nothing to help our situation and may even make it worse. Some of these may be serving as safety behaviours, others simply unhelpful ways to try to deal with a problem. In other words, the person’s learned coping responses are maladaptive. The aim of the therapy is to identify those maladaptive responses and replace them with a set of more adaptive ones. Somewhat related are approaches that focus on improving problem solving, particularly in the context of interpersonal difficulties. These focus on a broad approach of ﬁnding new ways to identify and understand the nature of a particular problem and alternative ways to resolve them. While both of these approaches include important cognitive elements and use cognitive techniques, probably the purest reﬂection of the cognitive approach are those that seek to identify and change the basic maladaptive thinking patterns that mediate the process, so-called cognitive restructuring approaches. We will mainly look at this latter approach as we go through this topic and the next. Slide 6 There are many ﬁgures that inﬂuenced the development of cognitive therapy over the years. The two most often credited with the development of the fundamental models and practice of cognitive therapy are the psychiatrist Aaron T Beck and the clinical psychologist Albert Ellis. Both were initially trained in psychoanalysis in the 1940s and ‘50s, when it was still the predominant therapeutic approach to the treatment of emotional disorders in America. Both came to reject the approach and were strongly inﬂuenced by the emerging evidence from cognitive psychology. For Beck and Ellis, the primary aim of their therapy was the reduction of emotional distress, whether depression, anxiety, anger, or other negative emotional state. Ellis was the ﬁrst, in 1957, to publish, teach, and apply a version of what we would now call cognitive therapy and which continues to be researched, taught, and used today. His therapy, initially called rational therapy, has many elements common to Beck’s slightly later cognitive therapy. Indeed, as the approaches have developed over the decades, the fundamental similarities greatly outweigh any differences. Ellis’s therapy was subsequently called rational emotive therapy and today has been broadened and is called rational emotive behaviour therapy. We will focus here on Beck’s cognitive approach, rather Transcripts by 3Playmedia Week © King’s College London 2018 2. than Ellis’s. This is not because it is fundamentally superior, but rather is a reﬂection of its greater inﬂuence that the work has had on subsequent research and practice. Both, however, share two core premises that underpin the cognitive models. The ﬁrst is that our emotional state is not a direct consequence of

WK 5 Combined Psychological Therapies PDF

Document Details

Tags

Related

Summary

Full Transcript

Upgrade to continue