44,95 €
An innovative new approach to addiction treatment that pairs cognitive behavioural therapy with cognitive neuroscience, to directly target the core mechanisms of addiction.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 369
Veröffentlichungsjahr: 2012
Table of Contents
Title Page
Copyright
About the Author
Preface
Chapter 1: The Tenacity of Addiction
Introduction and Overview
Discovering Cognition
Implicit Cognition and Addiction
Neuropsychological Findings
Addictive Behaviour is Primary, Not Compensatory
Changing Habits is the Priority
Diagnostic Criteria
Towards Integration
Equivocal Findings from Research Trials
Time for CHANGE
Evolution, Not Revolution
Something Old, Something New
Chapter 2: Existing Cognitive Behavioural Accounts of Addiction and Substance Misuse
The Evidential Basis of CBT for Addiction
Meta-analytic Findings
Behavioural Approaches
Diverse Treatments Mostly Deliver Equivalent Outcomes
What Are the Mechanisms of Change?
The Missing Variable?
A Dual-Processing Framework
Chapter 3: Core Motivational Processes in Addiction
Is Addiction About Avoiding Pain or Seeking Reward?
How Formulation Can Go Astray
Incentive Theories of Addiction
Learning Mechanisms in Addiction
Distorted Motivation and Aberrant Learning: the Emergence of Compulsion
‘Wanting and Liking’ in the Clinic
The Role of Secondary Reinforcers
Beyond Pleasure and Pain: a Psychoanalytic Perspective
Conclusion
Chapter 4: A Cognitive Approach to Understanding the Compulsive Nature of Addiction
Theories of Attention
Top-Down Influences Can Be Automatic
Automatic Processes Can Be Practically Limitless
Motivationally Relevant Cues are Prioritized
Biased Competition
Attention and Volition
Appetitive Cues Usually Win
Purposeful Behaviour Can Occur in the Absence of Consciousness
Attentional Bias and Craving
Cognitive Cycle of Preoccupation
Chapter 5: Vulnerability Factors In Addiction
Individual Differences in Addiction Liability
Personality Traits
The ‘Big Five’ Personality Factors
Personality Disorders
Affective Vulnerability Factors
Brain-Derived Neurotrophic Factors
Neurocognitive Vulnerability
Findings from the Addiction Clinic
From Research to Practice
Chapter 6: Motivation and Engagement
Impaired Insight and the Therapeutic Relationship
The Sad Case of Julia
Conflicted Motivation is the Key
Goal Setting and Maintenance
The Importance of Between-Session Change
Neurocognitive Perspectives on Motivation
Motivational Interviewing in Practice
Formulating and Planning the Intervention
Attributional Biases: the Blame Game
Case Formulation
Summary
Chapter 7: Managing Impulses
Introduction and Overview
Structuring the Session
Building Resilience
Impulse Control
Craving and Urge Report
Cognitive Processing and Craving
Cognitive Bias Modification
Attentional Bias in the Context of Addiction
The Alcohol Attention-Control Training Programme
Modifying Implicit Approach Tendencies
Reversing the Bias: Conclusion
Brain Training and Neurocognitive Rehabilitation Approaches
Clinical Implications of Delayed Reward Discounting
Tried and Tested Techniques
The Road to Recovery is Paved with Good Implementation Intentions!
Neurophysiological Techniques
Neuropsychopharmacological Approaches
Chapter 8: Managing Mood
The Reciprocal Relationship Between Mood and Addiction
Pre-existing Vulnerability to Emotional Distress
Negative Affect Due To Drug Effects
Stepped Care for Addiction
An Integrated Approach to Addressing Negative Emotion
Chapter 9: Maintaining Change
Relapse Prevention Strategies from a Neurocognitive Perspective
The Importance of Goal Maintenance in the Long Term
A Neurocognitive Perspective on Relapse
Twelve-Step Facilitation Therapy
Implicit Denial
Chapter 10: Future Directions
Neurocognitive Therapy
Increasing Cognitive Control is the Goal
Do We Know Anything New?
Appendix: Self-Help Guide Six TipsA Pocket Guide to Preventing Relapse
Introduction: Why Six Tips?
1. Don't Always Trust Your Memory!
2. Beware of the ‘Booze Bias’!
3. Separate Thoughts from Actions
4. Learn How to Distract Yourself
5. Willpower Is Sometimes Not Enough
6. Beware of the Dog that Doesn't Bark…
References
Index
This edition first published 2013
© 2013 Frank Ryan
Wiley-Blackwell is an imprint of John Wiley & Sons, formed by the merger of Wiley's global Scientific, Technical and Medical business with Blackwell Publishing.
Registered Office
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Offices
350 Main Street, Malden, MA 02148-5020, USA
9600 Garsington Road, Oxford, OX4 2DQ, UK
The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
For details of our global editorial offices, for customer services, and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell.
The right of Frank Ryan to be identified as the author of this work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
Library of Congress Cataloging-in-Publication Data
Ryan, Frank, 1944–
Cognitive therapy for addiction : motivation and change / Frank Ryan.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-470-66996-9 (cloth)—ISBN 978-0-470-66995-2 (pbk.)
1. Compulsive behavior–Treatment. 2. Substance abuse–Treatment. 3. Cognitive therapy. I. Title.
RC533.R93 2013
616.85′227–dc23
2012034393
A catalogue record for this book is available from the British Library.
Cover image: Die Furbige (The Intercessor) by Paul Klee, 1929. Photo © Geoffrey Clements / Corbis.
Cover design by Richard Boxall Design Associates
Dr Frank Ryan trained as a clinical psychologist at Edinburgh University and works as a consultant in Camden & Islington National Health Service Foundation Trust in London, UK. He practices as a cognitive behaviour therapist with a special interest in addiction and co-occurring disorders. He is an Honorary Senior Lecturer in the Centre for Mental Health, Faculty of Medicine at Imperial College and an Honorary Research Fellow at the School of Psychology, Birkbeck College, University of London. He is a former Chair of the Addiction Faculty of the British Psychological Society's Division of Clinical Psychology. He has also served as consultant in cognitive therapy to the United Nations Office on Drugs and Crime. The focus of his research is behavioural and cognitive processes in addiction and translating research into practice, with particular emphasis on findings derived from cognitive neuroscience.
The story begins with Bill, who was addicted to alcohol. He was attending a group along with eight other men and women in a specialist clinic in Hammersmith, West London, more than ten years ago. They also had experienced problems associated with their use of alcohol and were trying to abstain or reduce their level of alcohol consumption. As group facilitator, my first task was usually to ask members to ‘check in’ with an update on how the past week had been for them: the problems, the worries, the cravings, the lapses and the coping. When it was Bill's turn to say something about the week just passed, he froze momentarily. Unlike some in the group, Bill did not experience anxiety in social situations; on the contrary, he was usually a fluent, relaxed speaker. I asked whether he wanted to collect his thoughts and let someone else speak in the interim but he declined. He quickly recovered his composure and said that his pause was due to hearing the word ‘binge’ uttered by the woman sitting next to him. Bill apparently found this word distracting and he was unable to concentrate on what he had intended to say about his own ups and downs in the preceding week.
I was intrigued by this episode: if the mere mention of an alcohol-related word could be so distracting, how potent could other addiction-related cues be in capturing attention, especially outside the confines of the clinic, where temptation was everywhere? A subsequent literature search revealed just one study of what is termed attentional bias in addiction. This investigation (Gross et al., 1993) found that when cigarette smokers were deprived of nicotine for 12 hours they were more likely to be distracted by smoking related cues compared with fellow smokers who did not experience deprivation. Distraction was indexed by the slightly longer time it took the deprived smokers to name the colour used to print smoking-related words such as tobacco or lighter. Thus, they were slower to correctly respond with ‘red’ or ‘blue’ to these words than to neutral words such as locker or man. It was as if the words associated with cigarette smoking exerted a magnetic effect on the minds of abstinent smokers and distracted them from the primary task of simply naming a colour. The difference in reaction time between smoking-related and neutral words was tiny, a few milliseconds (ms), but was not observed with current smokers or people who had never smoked. To me, this appeared to be an analogue of what happened with Bill. Regardless of the task in hand, simply saying ‘red’, ‘blue’ ‘green’ or ‘yellow’ was slower if the word was connected with alcohol, but unaffected by the neutral words.
Although a definitive role for selective attention in anxiety disorders had by then been proposed (Williams et al., 1988), it was clear to me that attentional bias was equally important in relation to addictive disorders. The seminal work of Marlatt and Gordon (1985) had already highlighted the cue-specific nature of relapse in addiction, and how people could learn alternative coping strategies to forestall this. But what if an encounter with these so-called ‘high-risk situations’ reflected a cognitive bias rather than chance or circumstance? What if, after leaving the treatment centre or the rehabilitation unit, individuals were drawn to precisely the situations they were advised to avoid? Important questions, it seemed to me. But not just to me: cigarette smoking is estimated to cause 5 million deaths worldwide each year (Thome et al., 2009). In the United Kingdom in 2009, 8,664 deaths were attributed to alcohol-use disorders (ONS, 2011). It is estimated in the World Drug Report (UNODC, 2009) that between 11 and 21 million people in 148 countries worldwide inject drugs, of whom between 0.8 and 6.6 million are infected with human immunodeficiency virus (HIV). Addiction is also associated with massive healthcare costs: Gustavsson et al. (2011) estimated that in 30 European countries (27 European Union member states plus Iceland, Norway and Switzerland) addictive disorders cost €65.7 billion in direct and indirect healthcare costs. For comparison, anxiety disorders were estimated to cost €74.4 billion, and mood disorders (unipolar and bipolar depression) €43.3. An entire volume would be needed to describe the full extent of human misery and costs attributable to the spectrum of substance misuse and addiction.
Here, the focus is on the cognitive and motivational processes that enable diverse behaviours such as smoking a cigarette, sipping an alcoholic beverage or injecting heroin to persist in parallel with awareness of the harmful consequences that ensue and a sincere desire to desist. In order to learn more about the role of cognitive bias in addiction, I conducted an experimental study using a modified Stroop test (Ryan, 2002a) with the invaluable help of the clients and colleagues in the clinic. It seemed to me that if attentional bias could operate at an early stage of cue reactivity it could thereby influence the frequency and intensity of urges and clinical outcome as indexed by relapse rates. I began to explore the theoretical and clinical implications of this mental process, which seems to occur unconsciously, involuntarily and, by all accounts, relentlessly. This, I thought, helped to explain the disparity between the commitment to recovery shown by many addicted individuals and the high frequency with which they failed. It still seemed sensible to teach coping strategies in anticipation of encountering the people, places and things that might trigger appetitive impulses. But sometimes this seemed to be too little, too late.
By then, I realized that my interest in the role of cognition in addiction was shared by many talented researchers and clinicians. This helped me recognize that selective attention can only be understood, or at least partially grasped, when seen as a property of a highly sophisticated system of cognitive or executive control. Inspired by their efforts and continuing my clinical practice in parallel, I began to develop the ideas that form the basis of this book. These were elaborated through a series of presentations and workshops at events such as the European Association of Behavioural and Cognitive Therapies and the World Congress of Behavioural and Cognitive Therapies in exotic locations such as Acapulco, Vancouver, Paris and Dubrovnik. This entailed a reappraisal of cognitive therapy for addiction that accentuated the core theme of this text: addiction is quintessentially a disorder of conflicted motivation that is reflected in impaired cognitive control, defined as the ability to flexibly guide behaviour in the pursuit of desired outcomes or goals.
However, in the clinical arena within which many of the readers of this book operate, this cognitive–motivational process can often be obscured by the diversity of the presenting problems associated with addictive behaviour. Accordingly, it is necessary to place this focus on cognitive control in a broader therapeutic framework known as CHANGE, an acronym of Change Habits and the Negative Generation of Emotion. An acronym is always a compromise but CHANGE serves to remind those tasked with overcoming addiction, whether therapist or treatment seeker, that this entails reversing compulsive habits and managing emotions. The journey of the book thus began in the clinic, then detoured through a process of research and innovation only to return once again to the clinical arena. Along the way, academic and clinical colleagues have generously shared their knowledge and skills. I am deeply indebted to them. In particular, I would like to thank W. Miles Cox (who kindly commented on a draft of this book) Michael Eysenck, Matt Field, Hugh Garavan, John Green, Marcus Munafò, Mick Power, Anne Richards, David Soto, Philip Tata and Reinout Wiers. I remain, however, responsible for any shortcomings in the text! I am equally indebted to those who came my way in the clinic with insightful and authentic accounts of their own addictions.
Why does addiction exert such a tenacious grip on those who fall under its spell? In this book I propose that the answer to this question lies largely within the cognitive domain: the persistence of addiction is viewed as a failure or aberration of cognitive control motivated by the enduring and unconditional value assigned to substances or behaviours that activate neural reward systems. I shall outline how addictive behaviour endures because it recruits core cognitive processes such as attention, memory and decision making in pursuit of the goal of gratification, the associated alleviation of negative emotions, or both. This recruitment process is often covert, if not subversive, and operates implicitly or automatically in the context of impaired inhibitory control. The habituated drug user is effectively disarmed when exposed to a wide range of cues that generate powerful involuntary responses. The best, and often the only, option is to mount a rear-guard action from the command and control centre of the brain. This sets the scene for a reappraisal of cognitive therapy applied to addiction. Beginning with an overview of the plan and scope of the book, this introductory chapter outlines a cognitive perspective on addiction. It goes on to address shortcomings in historical and current therapeutic approaches to addictive behaviour and includes a brief review of the equivocal and occasionally puzzling findings generated in clinical trials. It concludes with an overview of CHANGE, the re-formulated account of psychological intervention based on cognitive, motivational and behavioural principles in a cognitive neuroscience framework that forms the basis of this text.
I have avoided the use of the term addict unless quoting from other sources. I do not think the manifestation of a particular behaviour should be used to denote an individual, in the same way that I would avoid use of terms such as a depressive or an obsessive in other circumstances. Of course, many of those who develop addictive disorders choose to refer to themselves as ‘addicts’. That is entirely appropriate for them, but I believe choosing to designate oneself as an addict is different from being so labelled by another. However, beginning with the title, I readily adopt the term addiction. Here, I apply a functional definition emphasizing the apparent involitional nature of addictive behaviour, its persistence in the face of repeated harm to self and others, and a tendency for drug seeking and taking to recur following cessation. In truth, addictive behaviour and its concomitant cognitive, behavioural and neurobiological facets occur on a continuum of varying, but often escalating, frequency and quantity or dosage. This is why attributing a static label such as addict is likely to miss the point, even if occasionally seeming to hit the nail on the head. There will be some interchange between the terms addiction, substance use and substance misuse according to the context. Generally, however, my use of the term addiction implies that the individual or group referred to meet standard diagnostic criteria for addictive disorders or dependence syndromes. Similarly, and again given pride of place on the front cover, I have opted for the term cognitive therapy rather than cognitive behavioural therapy. This decision is pragmatic rather than doctrinal but does authenticate the emphasis on cognition throughout the book. Both terms feature in the text, and anything deemed purely cognitive can easily be assimilated into the broader church of CBT.
Addiction has long been a source of fascination for theorists from a wide variety of scientific backgrounds. West (2001) listed a total of 98 theoretical models of addiction, which he classified broadly as either biological, psychological or social in orientation and content. Here, I do not attempt to review this diverse body of work. Nonetheless, West's taxonomy, referencing a ‘biopsychosocial’ framework, serves as a reminder that addiction is a complex, multifactorial, phenomenon. The main focus here is on understanding the neurocognitive and behavioural mechanisms of addiction and translating this knowledge into more effective therapeutic intervention. Most of the theoretical and empirical findings cited are based on either clinical trials or experimental paradigms involving drug administration, drug ingestion and drug withdrawal in humans and other species. For the most part, the substances at the root of the problems addressed in this text will therefore include opiates, cocaine, amphetamines, alcohol, nicotine and cannabis. At the time of writing, preparations for the fifth revision of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) are well underway. The term dependence, also central to the ICD-10 (WHO, 1992), is apparently being dropped. This is apparently due mainly to the possibility of conceptual confusion stemming from its dual meaning referring to either uncontrolled drug use, or normal neuroadaptation when, for example, narcotic analgesics are prescribed to alleviate chronic pain (O'Brien, 2011). The forthcoming taxonomy, due to be published in 2013, will therefore refer to ‘Addiction and Related Disorders’. Subcategories will refer to ‘alcohol use disorder’, ‘heroin use disorder’ and so on.
In the forthcoming diagnostic manual on addictive disorders, the chapter on addiction will also include compulsive gambling, currently classified as an impulse control disorder along with trichotillomania and kleptomania in DSM-IV (American Psychiatric Association, 1994). Consistent with this, Castellani and Rugle (1995) demonstrated that problem gambling is associated with tolerance, withdrawal, urges and cravings, high rates of relapse and high levels of co-morbidity for mental health problems. More fundamentally, from a cognitive neuroscience point of view, it is what goes on in the brain that matters, whether this is triggered by heroin, cocaine, alcohol or indeed gambling. By way of illustration, an intriguing series of case studies provides a more clinical dimension to the motivational power of dopamine, a key neurotransmitter in reward processing, in relation to gambling. Dodd et al. (2005) reported how they encountered 11 patients over a two-year period at a movement disorders clinic with idiopathic Parkinson's disease who developed pathological gambling. All of these patients were given dopamine agonist therapy such as pramipexole dihydrochloride. Seven of these patients developed pathological or compulsive gambling within 1–3 months of achieving the maintenance dose or with dose escalation. One 68-year-old man, with no history of gambling, acquired $200,000 of gambling debt. On cessation of dopamine agonist therapy his urge to gamble subsided and eventually ceased, an outcome also observed in the seven other patients that were available for follow-up. More generally, other behaviours with a propensity to become compulsive include online activities such as Internet addiction and gaming. My view is that a behaviour such as gambling that activates reward neurocircuitry with wins, and probably downregulates the same system with losses, is liable to become compulsive in susceptible individuals. Consequently, aspects of compulsive gambling and other behaviours where motivation to desist is compromised fall within the scope of this book.
The book begins with a brief critical appraisal of existing approaches, in particular cognitive and behavioural approaches such as cognitive behavioural therapy (CBT) and cognitive therapy itself (Chapter 2). This review is highly selective insofar as it focuses on shortcomings and unanswered questions, such as the finding that markedly diverse therapeutic approaches, including CBT, deliver broadly equivalent clinical outcomes. In successive chapters (3 and 4), I address first the core learning processes that contribute to the development of addiction and their neurocognitive bases, as well as delineating the predispositional role of exposure to adversity. Next, a conceptual framework that accommodates implicit cognitive and behavioural processes along with more familiar targets such as consciously available beliefs is outlined. The conclusion is that the most plausible way to regulate the former is by augmenting the latter: strategies that enhance executive control, metacognition or awareness are more likely to deliver better outcomes. By emphasizing a component process such as executive or ‘top-down’ control, the therapist and client are provided with a conceptual compass with which to navigate through the voyage of recovery. Chapter 5 addresses the question of individual susceptibility to addiction: if, indeed, drugs and gambling wins are such powerful rewards, why, ultimately, do not all but a small minority develop compulsive or addictive syndromes? This marks the transition from the more theoretical and research based chapters to content that is more directly relevant to the clinical or applied arena, although remaining grounded in a cognitive neuroscience paradigm.
Most of the remainder of the book (Chapters 6, 7, 8 and 9) explicates key therapeutic phases from a cognitive control standpoint. The sequence that unfolds follows the ‘Four M’ structure (see Figure 1.1), which is the enactment of the CHANGE approach:
Figure 1.1 The Four M Model. Clockwise, these are the four key stages.
Chapter 10 aims to summarize, integrate and look forward in the context of a vibrant research arena with major implications for the concept and conduct of cognitive therapy.
Existing accounts of cognitive therapy for addiction have not accommodated findings that cognitive processes, in particular those deemed automatic or implicit, are influential in maintaining addiction, or indeed as a potential means of leveraging change. In cognitive parlance, these models do not legislate for ‘parallel processing’ across controlled or automatic modes, with the latter being largely overlooked. Simply put, existing accounts fail to address what is the hallmark of addiction: compulsive drug seeking behaviour that appears to occur with little insight and often in the face of an explicit desire for restraint. Moreover, existing cognitive therapy approaches do not accommodate findings that cognitive efficiency is often impaired in those presenting with addictive disorders, whether stemming from pre-existing or acquired deficiencies. The client has developed a strong tendency for preferential cognitive processing and facilitated behavioural approach in the face of impaired cognitive control. Failure to acknowledge this leaves the therapist and client in the dark about an important source of variance that is influential at all stages of the therapeutic journey.
The findings of Childress and her colleagues (2008), who used advanced functional magnetic resonance imaging (fMRI) techniques to explore the neural signature of very briefly presented appetitive stimuli, are noteworthy. They found early activation of limbic structures such as the amygdala when the 22 participating abstinent cocaine addicts were shown subliminal, backward masked drug associated cues. A similar pattern was observed when covert sexual stimuli were presented. This design effectively eliminated the possibility of conscious recognition with backward masked exposure for a mere 33 ms, yet participants showed a clear pattern of activation in limbic structures implicated in reward processing. When tested with visible versions of these cues ‘off-magnet’ two days later, initial higher levels of brain activity in response to invisible cues was predictive of positive affective evaluation among the participants. As well as demonstrating the exquisite sensitivity of neural reward mechanisms to drug-related stimuli, these findings show that for habituated drug users the appetite for their drug of choice compares to powerful sexual drives: evidence perhaps that, for some, drugs are as good as, if not better than, sex. Further, Leventhal et al. (2008) found selective subliminal processing of smoking-related cues by nicotine-deprived smokers, again indicating non-conscious evaluative appraisal. It appears that, when exposed to significant cues, the brain makes up its mind very rapidly about what it wants. Extant theories (see, e.g., Marlatt, 1985; Beck, 1993) have difficulty in accounting for these cognitive events and processes, largely because information is processed at one level. Dual processing accounts, which form the basis of this text, have no such difficulty.
The definitive feature of implicit cognition is that ‘traces of past experience affect some performance, even though the influential earlier experience is not remembered in the usual sense—that is, it is unavailable to self-report or introspection’ (Greenwald and Banji, 1995, p. 4). These theorists illustrate the operation of implicit cognition with a generic example from experimental psychology. Participants are thus more likely to complete a word fragment or word stem using a word from a list to which they were previously casually exposed. Note that participants may not show explicit recall of the words but the effect of prior exposure nonetheless influences performance. The individual thus appears primed or predisposed to automatically generate a response that appears to evade introspection. This is, of course, precisely what is happening in the brains of the cocaine-addicted people referred to above: the drug-associated cues have acquired considerable emotional and motivational potency that assured them of preferential processing even the absence of conscious awareness.
In the addiction clinic, prior exposure to a vast array of appetitive stimuli, both focal and contextual, is the norm. Learning theory correctly charts the acquisition of conditioned behaviour, but is less able to accommodate cognitive processes, especially if these are implicit rather than manifest. Wiers et al. (2006) sought to clarify the scope of implicit cognition approaches in the addictive behaviour field by proposing three broad categories: attentional bias research, memory bias research and the study of implicit associations. Wiers et al. (2006) concluded that, at least in the populations of problem drinkers addressed in their article, there was an implicit bias towards the detection of alcohol-related stimuli. Following engagement of attention, subsequent information processing was shaped by implicit memory associations. Understandably, given their covert nature, these processes remain largely unseen and unheard by addicted people and their therapists. Moreover, their influence and expression is often masked in the sanitized environment of the treatment centre or clinic, thus creating a somewhat illusory sense of progress. For example, an individual who has just completed a detoxification procedure might explicitly predict their future progress, but implicit factors might improve predictive utility and thus influence the level and intensity of treatment subsequently received. Indeed, preliminary findings from Cox et al. (2002) indicated that alcohol-dependent patients who showed escalating levels of attentional bias to alcohol cues through the treatment episode were more likely to relapse. This raises the question of the feasibility and utility of modifying or reversing cognitive biases that will be addressed in Chapter 7. This finding was replicated by Garland et al. (2012), who found that attentional bias and cue-induced high-frequency heart-rate variability (HFHRV), assessed post treatment, significantly predicted the occurrence and latency to relapse at six-month follow-up in a sample of 53 people in residential care. This was independent of treatment condition (a 10-session mindfulness-based intervention and a comparable therapeutic support group) and after controlling for severity of alcohol dependence.
Implicit cognition might well be subtle but is also pervasive and can be detrimental for both therapeutic engagement and clinical outcomes. Accordingly, cognitive therapy needs to accommodate a broader concept of cognition in addiction, delineating a role for implicit processes in parallel with the more familiar focus on conscious deliberation and re-appraisal. This re-conceptualization is the basis for developing the innovative approaches to formulating and intervening with addictive and impulsive appetitive behaviours that will be addressed in this text. The theoretical framework and clinical strategies are thus derived from CBT but framed within a cognitive neuroscience paradigm. I shall describe how this emergent paradigm can augment existing therapeutics and also generate innovative techniques that directly target the core cognitive and behavioural mechanisms of addiction.
In the context of overcoming addiction, cognitive control is concerned with maintaining recovery goals and monitoring progress in goal pursuit. In particular, managing addictive impulses that have become redundant, unwanted or risky is vital. Cognitive control, especially inhibition, thus forms a key component of the broader executive functioning necessary for self-regulation. Other components of this function, associated with the prefrontal cortex, include shifting strategies in response to changing task requirements and updating by monitoring of goal pursuit. These cognitive operations—shifting, inhibiting and updating—have emerged as relatively independent factors in experimental investigation of executive functioning Miyake et al., (2000). Impaired control over drug use by habituated users is of course a definitive feature of substance dependence and thus a rather obvious target for therapeutic intervention. Cognitive neuroscience findings provide confirmatory evidence for this. Chambers et al. (2009) reviewed evidence pointing to cocaine users, for example, showing impairments on several laboratory measures of impulse control such as having to withhold a well practised response or manifested in making riskier decisions. These deficiencies have been noted both under conditions of acute intoxication and also among abstinent restrained drug and alcohol users. Kaufman et al. (2003), for instance, using fMRI during a go/no go task, found significant cingulate, pre-supplementary motor and insular hyperactivity in a sample of 13 active cocaine users when compared with 14 cocaine naive controls. Forman and colleagues (2004), also using fMRI found abnormally decreased activity of this cingulo-frontoparietal–cerebellar neural network with a cohort of opiateaddicted individuals. These findings suggest that some drugs with addiction liability compromise command and control centres in the brain. This happens during intoxication but also carries over to abstinent periods. The most important faculty required for recovery, cognitive control or willpower, is thus rendered less effective when it is most needed.
Conversely, clinical neuropsychological findings indicate that damage to cortical structures can dramatically disrupt appetitive behaviour. Yucel et al. (2007) used a combination of fMRI and proton magnetic resonance spectroscopy to investigate cognitive control in a cohort of 24 opiate-dependent individuals on either methadone or buprenorphine with drug-naive controls. They found that, while anterior cingulate cortex (ACC) activation was equivalent across the two groups, the opiate users failed to demonstrate the normal association between ACC physiological activity and behaviour measures (i.e. response errors) shown by the control group. There were abnormalities detected in neurochemical markers such as N-acetyl aspartate and glutamate. In addition, opiate users required greater involvement of the frontoparietal and cerebellar behavioural regulation network to achieve normal levels of cognitive control. Yucel et al. (2007, p. S99) speculated that ‘The pattern of results across these studies implies that chronic drug use leads to the recruitment of a compensatory network of brain regions in order to successfully detect and resolve conflicts in response tendencies. However, even though normative behavioural performance may be achieved in structured laboratory experiments, the same neural systems may be more vulnerable to fail in the real world, where emotional and motivational influences (e.g. stress, craving, withdrawal, etc.) also tax these cognitive and neural resources’. Interestingly, both the clinic cohort and matched controls showed robust neurocognitive functioning as indicated by obtaining IQ scores of 112, above the normal range. The influence of opiate substitution therapy (average methadone, about 43 mg; average buprenorphine, 10 mg), or indeed premorbid cognitive performance deficits, cannot of course be ruled out as a source of differences between the clinic attendees and the control group. However, in the studies reviewed here, while participants from both experimental and control groups made errors such as pressing a button when they did not intend to, it was those with the addictive history who were somewhat underwhelmed at the neurobiological level, as evidenced by hypoactivity in the cingulate.
With regard to other substances, heavy marijuana use was associated with lower performance on tests of memory, executive functioning and manual dexterity in a sample of 22 regular users who had remained abstinent for 28 days prior to testing (Bolla et al., 2002). Eldreth et al. (2004) compared performance on a modified version of the Stroop task and brain activity between 11 heavy marijuana users who had been abstinent for 25 days and 11 matched comparators. The marijuana users showed no comparative deficits in performance on the Stroop task. However, the marijuana users showed hypoactivity in the ACC and the left lateral prefrontal cortex and hyperactivity in the hippocampus bilaterally, a pattern not observed in the comparator group. This suggests that the marijuana-using group was relying on a compensatory mechanism in the face of suboptimal error monitoring.
Turning to neurocognitive deficits associated with cocaine use, Bolla et al. (2004) used a variant of the Stroop test and positron emission tomography to explore cognitive conflict in 23-day-abstinent cocaine users. While engaged on the Stroop task, cocaine abusers showed less activation than non-drug-using comparison subjects in the left ACC and the right lateral prefrontal cortex but greater activation in the right ACC. This pattern was associated with response competition generated by the Stroop task. Resting scans showed no differences in neural circuitry such as the ACC, which subserves executive functioning. The abstinent cocaine users with the most intense of history of drug use showed the highest level of abnormality when cognitively challenged. Importantly, the two groups performed at equivalent levels on the cognitive tasks, being differentiated only at the neuronal level.
How should these findings be interpreted in the context of responding to addictive behaviour in the clinical arena? First, it is neither possible nor necessarily helpful to speculate on the issue of causality. In this regard, Garavan and Stout (2005) hypothesized that observed functional deficits such as those briefly reviewed above could be the product of pre-existing cognitive ‘trait’ variables such as poor impulse control that could be potentiated by acquired patterns of substance misuse, thus leading to more transient ‘state’ variables such as compromised cognitive control associated with recent intoxication. The fact that these deficits appear to endure for at least a month after cessation of drug use suggests that those involved in delivering therapeutic intervention should, at the very least, be aware of the fact that the client may be compromised in terms of cognitive control or at least have to exert more mental effort in dealing with situations where rapid decision or response inhibition is required.
As evidenced by the robust performance on neuropsychological tests, there did not appear to be global differences in cognitive functioning across the various groups of substance using and control participants. Instead, there appeared to be a specific deficit revealed by precise cognitive tests and neurobiological assays among the cohorts of drug users recruited. From the standpoint of embarking on a journey of rehabilitation that will inevitably involve complex new learning, it appears to me that a deficit in detecting errors or perhaps the necessity to exert more cognitive effort to compensate for this will inevitably prove challenging, if not exhausting. Moreover, these subtle alterations in cognitive processing remain largely unrecognized outside the cognitive neuroscience laboratory. At the very least, awareness of this should enable the addiction therapist to generate more accurate empathy as their clients encounter the inevitable challenges on the route to recovery.
Garavan and Stout (2005) concluded on the basis of their review that drug misusers evidenced a low level of awareness of errors on a range of laboratory tasks. This was associated with hypoactivity in the ACC, a structure vital to performance monitoring . The ACC is the early warning system for errors and is activated during action slips (Garavan et al., 2003), and contributes to the ‘D'oh!’ feeling epitomized by Homer Simpson when he makes yet another error. In a sense, these findings validate phenomenological aspects of drug use. When clients state that they are struggling to cope with the challenges of restraint as they work towards recovery, the informed therapist can thus provide a more empathic response. As will be seen, recognizing deficits in cognitive control is the first step in developing the emerging ‘neurocognitive’ therapy addressed as outlined in Chapter 6 and elsewhere in this volume. This focus on cognition is thus intended to create a therapeutic space that bridges the gap between the neurobiological mechanisms of addiction and the need to devise plausible therapeutic strategies. The rationale is that CBT can be more accurately formulated and precisely targeted by understanding the enduring neurocognitive signature of addiction.
Second, existing or historical accounts emphasize or assume that addiction is compensatory: compulsive drug use and gambling are seen as a means of dulling or avoiding emotional pain, rather than seeking pleasure or reward. This follows a long tradition in psychology, largely unfettered by empirical support. In the psychodynamic tradition, Kohut (1971, p. 46), for example, viewed drug use as ‘a replacement for a defect in the psychological structure’. The major failing of intuitive accounts of addiction such as this was circular reasoning. They echo historical accounts proposing that ‘moral deficiency’ was the cause of addiction on the decidedly shaky premise that individuals were lacking moral fibre simply because they were addicted. In turn, the addiction was paraded as evidence of the underlying moral frailty.
More recently, cognitive therapy accounts have similarly proposed addiction as a reaction to an event or an emotional state rather than a primary motive. Addiction stemmed from attempts to cope with or suppress maladaptive core beliefs such as ‘I am helpless’ or ‘I am unlovable’ (Beck et al., 1993, p. 52). Cognitive therapists reading this text will doubtless have elicited these beliefs in the course of their work in addiction clinics and elsewhere. Clearly, negative emotions such as depression and anxiety can lead to drug taking, and dysphoria can be also be consequential to intoxication. However, vulnerability to emotional disorders and addictive disorders could also emerge and present in parallel because of their prevalence: lifetime prevalence rates for anxiety disorders in the USA, for example, have been estimated at 29% and mood disorders at 21% of the population (Kessler et al., 2005) and an estimated 120 million of the adult population regularly consume alcohol (Anthony et al., 1994).
It seems plausible that some individuals with coexisting emotional and addictive problems could have acquired these through different mechanisms or learning processes. In this regard, Hiroi and Agatsuma (2005) reviewed evidence indicating genetically distinct pathways leading to expressions of either drug dependence or comorbidity. Presumably, these could coexist. For instance, an individual could acquire an anxious disposition through a combination of genetic predisposition and exposure to adversity in childhood. The same individual could develop a dependence on cocaine or alcohol because of a different combination of genetic predisposition and an environmental factor such as easily available cocaine and hence greater exposure. Significantly, when cigarette smokers speak of their addiction, the listener rarely responds with speculation that this reflects some compensatory behaviour linked to disrupted attachment to parents, or being the victim of cruelty as a child. The addicted smoker is implicitly viewed in a manner more akin to that proposed in this text: regardless of his or her pre-existing vulnerabilities, it seems clear that the problem is inability to give up smoking. Similarly, a lifelong heroin user, recently detoxified, said to me in the course of a brief screening assessment: ‘Basically I'm fine; I'm just an addict’. He did in fact appear to enjoy a sense of well-being. Many of course do not, and this can lead the formulation astray as the therapist and client strive to connect up the addictive behaviour with the negative emotional legacy. Generating a valid conceptualization of addiction is crucial if cognitive and cognitive behavioural approaches are to deliver enhanced outcomes. In the absence of conceptual clarity, confusion can thus emerge. Consider this statement from a client who resumed using cocaine and drinking alcohol after eight months abstinence:
What is it about me, just when I get things right for a change, I start using again and end up relapsing. I seem to push the ‘self-destruct button’. It must be that, deep down, I just want to be a failure (Ryan, 2006, p. 291).
Cognitive therapists (at least this one!) would most likely need to engage their own inhibitory systems to avoid tackling this maladaptive core belief. The CHANGE model does not ignore the possibility that such core beliefs are therapeutically significant, but questions whether they should be the primary focus. Here, a more parsimonious account of addictive behaviour is offered. The attention of the therapist and client when reviewing this episode focused on the more proximal antecedents of behaviour rather than searching for underlying vulnerabilities.
Here, addiction is viewed as initially appetitive or hedonic, at least in the acquisition phase, when neural reward systems are first transformed by repeated drug ingestion. The hedonism or pleasure may well diminish with habituation but the approach behaviour remains as compelling as ever. It therefore contrasts sharply with common mental health problems such as anxiety and depression manifested (and maintained) by avoidance behaviour, or effortful suppression of unwanted thoughts or images. Existing cognitive therapy accounts of addiction (e.g., Beck et al., 1993) view stress or interpersonal conflict as contexts for eliciting beliefs such as ‘A drink will relax me’, which in turn elicit automatic thoughts such as ‘Drink!’ or ‘Smoke!’. These thoughts evoke craving and urges to use drugs. Without doubt, this is a potential pathway to a lapse or relapse, and a competent therapist would not hesitate in addressing this potentially maladaptive sequence of thought and action. Here, I propose that, while virtually any event or situation can become a precursor to drug use, there is a more direct cognitive–motivational process activated when drug cues are detected. Moreover, negative affect is by no means the only pathway, and positive affect or factors such as testing personal control also appear to precipitate relapse. Granted, these lapses did not prove as sustained as those associated with negative affect (Hodgins et al., 1995). Nonetheless, a different picture emerged when a prospective design was employed to investigate the affective antecedents of relapse among a group of 133 smokers. Shiffman and Waters (2004) found that more enduring, day-to-day changes in stress and negative affect appeared to have little influence on lapse risk, but that more sudden increases in negativity, perhaps triggered by an argument and lasting for shorter periods of hours or minutes, were more likely to promote smoking lapses. They concluded that smokers aiming to give up are best advised to learn to cope with the challenges posed by the transient ‘slings and arrows’ of everyday life. Changes in affect, of whatever valence, are dynamically linked to addictive behaviour but are neither necessary nor sufficient to account for its genesis and enduring legacy. On occasion affective changes serve as triggers for compulsive drug taking or gambling, but often no excuse is needed. Addictive pursuits are intrinsically rewarding.
Therefore, in the current text I am according more immediacy and primacy to drug cues as powerful and direct