Pain as a Clinical Problem

Commentary: Breaking Tradition and Breaking New Ground: The Contributions of WE Fordyce to the Evolution of Pain Theory and Practice

Michael J.L. Sullivan, PhD and Michael Nicholas, PhD

In this chapter, Fordyce lays out a conceptual framework for understanding psychological contributions to pain. On re-reading Fordyce’s book 37 years since it was first published, one of the most notable features is that many of the debates that rage in pain science and clinical practice today were also vibrant back in the 1970s.

In Chapter 1, Fordyce was concerned with setting the conceptual framework for a behavioral perspective on chronic pain and the key debates centered around ideas on the relationship between pain experience and activities in the nervous system, as well as the role of psychological factors in explaining both the experience and impact of pain on the individual patient.


One of Fordyce’s first challenges is to make a case for the importance of psychological influences on pain experience and pain behavior. Today, such a case easily could be made by appealing to a vast literature highlighting how a wide range of psychological factors play a role in the experience and expression of pain. However, in 1976, to argue for greater attention to the psychology of pain would have been a daunting task. In 1976, empirical research on the psychology of pain was only in its infancy and clinical lore dominated discussions on the topic.

Although the role of psychological factors in pain had been addressed by several authors before Fordyce, explanations of psychological influences on pain had proceeded primarily from psychoanalytic or psychodynamic models. In 1959, Engel had published his influential paper, Psychogenic Pain and the Pain-Prone Patient [8]. The notion put forth in the paper was that certain personality features could predispose individuals to the development of pain symptoms. Faced with a situation where patients presented with physical symptoms with no discernible organic pathology, to the medical practitioner of the day, the “leap to the head” appeared like a reasonable approach to understanding the pain patient. Unfortunately, many of the psychodynamic variables implicated as causative of chronic pain were poorly defined, impossible to measure and had minimal value for treating chronic pain [29,38].

Fordyce takes steps to distance his proposed conceptual framework from his psychodynamic predecessors. First, he addresses the issue of variability in pain experience and pain expression that is often evidenced in patients with chronic pain. In essence, the variability in the clinical presentation of chronic pain was often used as a basis for arguments concerning the “non-medical” origin of chronic pain. Fordyce suggests that environmental/social contingencies that differentially reinforce pain behaviors across situations could explain variability in symptom presentation. Fordyce further argues that the influence of environmental/social contingencies did not question the authenticity of the patient’s underlying clinical condition.

Fordyce utilized the writings of Clark and Hunt [3] to frame the philosophical challenges represented by mind-body dualism. In this case, Fordyce questioned why events taking place in the nervous system often seemed to be construed as somehow more important than the experience of the patient and the factors contributing to the impact of the pain on that person’s day to day behaviors. Fordyce observed that “pain is not an entity or thing; it is a label that observers, including the pain sufferer, have attached to relevant phenomena they have observed or experienced: [p. 19]. Fordyce argued that limiting explanations of pain phenomena to neurophysiology seemed unnecessarily restrictive and risked overlooking many other contributors that could be of therapeutic benefit.

To emphasize this point, Fordyce recounted a story told by the comedian Robert Benchley about an economics student who wrote an exam essay on economic factors affecting the fishing industry by discussing the topic from the perspective of the fish. Accordingly, Fordyce argued that chronic pain needs to be viewed from the point of view of the person in pain (the fish), rather than being considered a collection of neurophysiological events. He did not deny the importance of appreciating pain from a neurophysiological perspective, just that this perspective had become too strongly emphasized, often at the risk of detriment to the pain patient–for whom suffering and disability were effectively regarded as unfortunate by-products or epiphenomena. Fordyce argued that while neurophysiological accounts of pain phenomena should be acknowledged and incorporated into a comprehensive explanation for pain, too much emphasis on neurophysiological mechanisms risked many pain problems being effectively dismissed, especially when no clear ”organic” basis for a patient’s pain report could be identified. In these cases, it has always been tempting to attribute the patient’s complaints as due primarily to psychologica’ factors – the psychogenic thesis. This either organic or psychological perspective had been strongly debated well before the 1970s, but remained potent until relatively recently with the terms “medically-unexplained symptoms” and “somatization” being widely used as pseudonyms for “psychogenic” (see [25] for a summary of this debate). More recently, Flor and Turk [9,10] provide a similar echo for Fordyce’s position. These authors describe many of the developments in our knowledge since the 1970s of the joint contributions that biological, psychological, and environmental factors can make to an individual chronic pain patient’s experiences.

Instructively, the 2013 edition of the American Psychiatric Association’s Diagnostic and Statistical Manual (DSM-V) has eliminated any reference to somatization [1]. The authors of the DSM-V argue that medically unexplained symptoms are present in many diagnosed medical disorders, and they consider it inappropriate to give a person a mental disorder diagnosis just because a medical cause could not be demonstrated. Fordyce would have whole heartedly agreed.


In order to make the case for the importance of psychological influences on pain experience and pain expression, Fordyce also addresses the limited explanatory power of physiological models of pain that had been proposed to date. This he does by providing a brief historical account of Von Frey’s Specificity theory [35] and Goldscheider’s Pattern theory [15]. Specificity theory held that the nervous system contained specific pain receptors that yielded pain sensation upon stimulation. Pattern theory held that patterns of stimulation (e.g., distribution, intensity), even in non-specific receptors could be ‘coded’ at the periphery as nociceptive information leading to a sensation of pain. Fordyce argued that research findings showing that neural ablations did not necessarily relieve pain, that emotional factors influenced pain perception, and that there existed cultural variations in pain expression presented significant challenges for specificity and pattern theories.

Fordyce applauds Melzack and Wall’s Gate Control Theory (GCT) [24] as representing a significant advance over specificity and pattern theories. The GCT was able to account for variation in pain symptoms as a function of a gating system that modulated nociceptive information in the central nervous system. Importantly, the gating system was under partial cortical control, thus providing a framework for understanding psychological influences on pain.

The GCT in many ways revolutionized pain research. Not only did the theory propose a multidimensional conceptualization of pain that included psychological processes, it helped explain clinical pain phenomena such as injuries without pain, and pain that existed in the absence of a discernible lesion [37]. While the GCT had greater explanatory power than specificity or pattern theories, Fordyce highlighted that the GCT was still a theory about pain sensation and did not address the behavioral dimensions of pain. The GCT did consider a place for behavior, but the ‘action system’ of the GCT operated at the spinal level; the role of behavior was relegated to the domain of reflexes. Research prompted by the GCT addressing the ‘action system’ of pain focused primarily on animals, and the actions studied were reflexive withdrawal or escape responses [24,36].

Decades later, Melzack [23] would propose a ‘neuromatrix’ model of pain, which greatly expanded the dynamic role of networks within the brain to explain the experience of pain. According to the neuromatrix theory, the brain has a neural network that integrates information from multiple sources and levels to produce the sensation of pain. Melzack [23] suggested that various brain regions produce cyclical processing and synthesis of nerve impulses through the neuromatrix and give rise to a “neurosignature” of pain. Albeit a fascinating extension of GCT, beyond reflexive withdrawal or escape, the neuromatrix model was still silent on the role of behavior in the pain system.


Fordyce suggested that pain was not only something that affected what people feel, pain also affected what people do. A wide range of movement alterations can be observed in individuals experiencing pain. These might include facial or postural configurations, actions oriented toward protection or care of an injury, and actions oriented toward the pain-related stimulus or pain-relevant environment. From Fordyce’s perspective, any phenomenon that had a behavioral dimension could fall under the control of environmental contingencies.

Proceeding from principles of operant conditioning, Fordyce suggested that pain behaviors, like any other behaviors, were influenced by their consequences. A particular behavior (e.g., moaning) that is followed by a positive consequence (e.g., empathic attention) will have a higher probability of being emitted in the future, regardless of the ‘level’ of pain. In this case, ‘moaning’ becomes instrumental in achieving empathic attention. Of course, at the time he was outlining this behavioral perspective of pain there was little direct experimental support for these propositions. However, a series of studies based on a paradigm first described by Linton and Gotestam [21] have since provided strong experimental support for the proposition that pain report (pain intensity ratings) can be seen as a form of behavior susceptible to reinforcement contingencies independent of noxious input [18,22]. It followed that treatments that manipulated reinforcement contingencies might hold promise of reducing pain-related behaviors.

The focus of Fordyce’s approach to treatment was not on reducing the experience of pain directly, but on reducing the overt display of pain. The targets selected for treatment were pain behaviors such as distress vocalizations, facial grimacing, limping, guarding, and medication intake.

The first behavioral approaches to the management of pain and disability were conducted within inpatient settings that permitted systematic observation of pain behaviors, as well control over environmental contingencies influencing pain behaviour [11]. Staff members were trained to monitor patients’ behavior, and to selectively reinforce “well” behaviors and selectively ignore pain behaviors [14]. Results of several studies revealed that the manipulation of reinforcement contingencies could exert a powerful influence on the frequency of display of pain behaviors [13]. The manipulation of reinforcement contingencies was also applied to other domains of pain-related behavior and shown to be effective in reducing medication intake, reducing downtime and maximizing participation in goal-directed activity.

A number of clinical trials on the efficacy of behavioral treatments for the reduction of pain and disability yielded positive findings [28]. However, given the significant resources required to implement contingency management interventions, issues concerning the cost-efficacy of behavioral therapy for pain and disability were raised. Concern was also raised over the maintenance of treatment gains since reinforcement contingencies outside the clinic setting could not be readily controlled. In order to increase access and reduce costs, behavioral treatments were modified to permit their administration on an outpatient basis. This change in delivery format compromised to some degree the control over environmental contingencies, and required greater reliance on self-monitoring and self-report measures [28].

The early work of Fordyce and his colleagues was at once novel and contentious with its focus on reducing pain behavior as opposed to ameliorating the pain condition. Critics voiced their concerns that behavioral treatments might only be effective in training stoicism, and were not dealing with the underlying problem. In response to critics, Fordyce pointed to literature indicating that the magnitude of the relation between pain and disability was modest at best, and that treatments aimed at reducing pain often had no appreciable effect on level of disability [13]. According to Fordyce, to effectively treat disability, the disability had to be targeted directly [12].


Fordyce’s seminal work has continued to influence research and theory on behavioural dimensions of pain. Current fear-avoidance models of pain and disability were inspired by Fordyce’s behavioural approach to treating chronic pain. Vlaeyen and his colleagues have proposed that pain-related fears can become associatively linked to activity (or the anticipation of activity) and contribute to avoidance behaviour [32]. Individuals with musculoskeletal pain conditions typically experience increases in pain following activity, and decreases in pain upon activity cessation. This relation between activity and pain sets a stage ideal for the learning of escape or avoidance behavior [33]. Activity cessation or activity avoidance is reinforced by the reduction in experienced pain. Since activity cessation and activity avoidance essentially define “disability’ ” learning principles have been discussed as fundamental to the development of disability associated with pain [34, 35]. It has been suggested that disability can persist solely on the basis of an individual’s expectation that a certain behavior will result in pain [20]. Since avoidance of activity does not allow for disconfirmation of the individual’s expectation or belief that pain will result from activity, disability can be maintained long after a pain condition has resolved.

Proceeding from a strict behaviorist tradition, Fordyce made no assumptions about the functional basis of pain behavior. More recently, psycho-evolutionary models have been put forward suggesting that pain behaviors may be integral components of the pain system and are adaptive insofar as they enhance survival potential [30]. Within these models, a distinction has been drawn between communicative and protective pain behaviors. Communicative pain behaviors include facial expressions such as grimacing or wincing, and verbal or paraverbal pain expressions such as pain words, grunts, sighs, and moans. The overt display of distress during pain experience conveys information to observers about the internal state, pain-related limitations, and needs for assistance of the individual who is experiencing pain [5,7,17,39]. Protective pain behaviors include any action that is intended to reduce the probability of further injury, minimize the experience of pain or promote recovery from injury. For example, the withdrawal of a limb from a hot surface serves to terminate the action of a noxious stimulus and in turn, protects the limb from further injury [36]. Similarly, the use of limping to alter weight distribution during ambulation might minimize pain to an injured limb and reduce the probability of injury exacerbation [4,6]. Protective pain behaviors also include actions such as guarding, holding the injured or affected area of the body to reduce movement that might be associated with pain [19].

It has been suggested that feelings of empathy or sympathy triggered by the observation of others’ experience of pain can prompt caregiving or support behavior [5,16,39]. Social behavioral responses might promote survival by providing critical care for an individual who is ill or injured and can also act as potent reinforcers influencing the probability of the future occurrence of pain behavior [11,27].

Beyond their survival value, it has been suggested that expressive pain displays could also be used strategically to induce others to alter their expectations, reduce the performance demands placed on the individual in pain or as a means of managing interpersonal conflict [31]. Psycho-evolutionary and learning models of pain behavior are not incompatible but they differ with respect to the characterization of the individual as a passive respondent to reinforcement contingencies or an active engineer of social geography during times of distress.


Perhaps a more challenging proposition for many modern readers would be Fordyce’s assertion that “pain is not simply what a patient says it is” [p. 20]. This is a position that many writers have assumed. For example, in the notes attached to the IASP’s [26] widely-accepted definition of pain it is stated: “If they regard their experience as pain and if they report it in the same ways as pain caused by tissue damage it should be accepted as pain.” The intention of this perspective was to avoid disputes over whether a person was or was not experiencing pain and to shift the focus to investigating possible explanations for the reported experience. While Fordyce would, no doubt, have concurred with this intention, his point was that first, an individual may not have a full appreciation of all possible influences on his/her experiences and second, that a person’s report of pain remains a form of behavior, and like all behaviors there may often be a poor correlation with other behaviors or bodily states. He likened this to the idea that there is often a discrepancy between what someone may say and what they may do. Fordyce utilized the operant concept of different reinforcement contingencies (or consequences) between different behaviors leading to differential learning as a possible explanation for these disparities. To illustrate his case, Fordyce used the commonly observed disparity amongst cigarette smokers who say they are strongly committed to quit smoking and continue to smoke. The key point here is that all behaviors occur in a context (in this case, a specific learning history) and these contexts are likely to vary greatly between behaviors. Thus a person’s report of being in pain or their display of pain behaviors (like grimacing or limping), could well reflect their previous experiences (and consequences) of such behaviors, rather than a particular experience that would meet the IASP’s definition of pain.

Fordyce summarized this point by saying, “The scope and definition of a clinical pain problem is to be found in what the patient does as well as what he or she says. Patient behavior is a critically important element” [p. 22]. It is interesting to note that this perspective was never taken up by the authors of the IASP definition of pain even though it preceded the IASP publication by 3 years and the IASP definition has not been revised since its initial publication. Perhaps this review of Fordyce’s contribution to pain science and clinical practice might finally lead to a review of the IASP definition as well.

As mentioned earlier, unlike many of the early behaviorists, Fordyce did not consider the nervous system as a type of “black box” that should be acknowledged but overlooked in favor of the study of stimulus-behavior relationships. In Chapter 1, he outlined the main neurophysiological accounts of pain to that time, including the relatively recent development of the gate control theory (GCT). But more importantly, he went further than mere descriptions and also considered their implications for understanding and treating chronic pain. In this regard, his recognition that the GCT, for all its conceptual significance and relevance for acute pain, could not adequately account for the phenomena of chronic pain (and offered little help in treating chronic pain) was ahead of his time. As he pointed out, any therapeutic strategies based on the GCT would necessarily be time-limited and difficult to sustain. Indeed, it was not until the mid to late 1990s that Melzack and colleagues recognized this limitation in the original GCT with their development of the terms “neuromatrix” and “neurosignature” to explain how the experience of persisting pain might be maintained independently of noxious/sensory input. More recent research has, of course, identified multiple neuroimmune interactions following injuries to both central and peripheral nerves as another possible neurophysiological basis for some chronic pain conditions [40]. Other basic science researchers have described evidence suggesting that classical conditioning processes play a key role in the development and maintenance of chronic pain [2]. Interestingly, in all these evolving neurophysiological accounts of chronic pain, the treatment methods described by Fordyce throughout his book, based on a behavioral analysis of chronic pain, can be considered to have remained relevant to clinical interventions. In this sense, one can see a fundamental feature of Fordyce’s introductory chapter was his ability to construct a coherent theoretical perspective that allowed for developments in fields like neurophysiology and psychology, without becoming exclusive. Subsequent developments in psychological pain research may have greatly extended our understanding of chronic pain and multiplied the range of treatments subsumed under the rubric of “psychology,” but the key points outlined by Fordyce in Chapter 1 can be seen to stand true even 37 years later.

Key points:

1.  Pain is not a discrete entity but a complex network of phenomena about which there is a broad range of perspectives. Approaches to the subject must be prepared to identify which facets are being dealt with and to recognize that others may view the problem differently.

2.  There is but a loose linkage between noxious stimulation peripheral to the central nervous system and sensations presumed to emanate from this stimulation, even when viewed in strictly neurophysiological terms.

3.  Previous experience and ongoing cortical activity are ensured a role in even the simple detection of a pain stimulus, to say nothing about how the sensation is perceived and what responses ensue.

4.  The overt dimensions of pain (e.g., pain behavior) are likely controlled by environmental contingencies that bear little or no relation to the neurophysiological processes underlying more covert (e.g., pain sensations) dimensions of pain.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 15, 2018 | Posted by in Uncategorized | Comments Off on Pain as a Clinical Problem

Full access? Get Clinical Tree

Get Clinical Tree app for offline access