Although clinical studies and animal models seem to establish an important role for the sympathetic nervous system in many forms of neuropathic and inflammatory pain, there is an ongoing debate on the classification of pain syndromes with sympathetic components. The confusion originates from several sources: failure to acknowledge that the pathomechanism of chronic pain can change during the progress of the disease, which is now strongly underlined by experimental data from suitable animal models. Neuropathic pain is a vaguely defined collection of pain syndromes which includes painful conditions with diverse and largely unknown patho-mechanisms. Clinical diagnosis is difficult and well designed, placebo controlled sympathectomy is rarely performed. [blumberg et al.]

The brain seems to be able to generate and uphold sensitization by itself, based on previous experience, or genetic disposition. This seems to be particularly important for muscle pain. There seem to be positive feedback loops where pain produces more pain, and more sensitization. Musculoskeletal pain is the most common pain state. It amounts to almost 50% of all long term sickness absence. But other subjective complaints are also common, and may depend on sensitization. Sensitization has been introduced as an explanation for subjective complaints from the gastrointestinal tract and the brain, like fatigue, tiredness, dizziness, and vertigo. [coderre & katz]

This commentary on coderre & katz, wiesenfeld-hallin et al., and dickenson focuses on: (a) the brain as an under-recognized contributor to pain facilitation at the spinal cord; (b) these brain-to-spinal pathways being activated by learning or by body infection/inflammation; and (c) the resultant spinal release of anti-analgesic neuropeptides, activators of the NMDA-NO cascade, and activators of glia.

Intradermal capsaicin in humans causes pain, primary hyperalgesia, and secondary mechanical hyperalgesia and allodynia. Parallel changes occur in the responses of primate spinothalamic tract cells and in rat behavior. Neurotransmitters that trigger secondary mechanical hyperalgesia and allodynia include excitatory amino acids and substance P. Secondary mechanical allodynia is actively maintained by central mechanisms. Our group has investigated mechanisms of central sensitization of nociceptive neurons by examining the responses to intradermal injection of capsaicin. These experiments are pertinent to issues raised by coderre & katz (sect. 2).

Despite significant advances in pain research and clinical pain management, little effort has been devoted to exploring whether the same pain treatment strategies are effective for male and female patients. Recent studies indicate that sex differences might play a role in the response to noxious events and in the response to analgesic interventions (berkley). Further insight into these gender differences will lead to improved pain management for women and men.

Sex is one of biology's, that is, life's most potent experimental variables. So, are there sex differences in pain? And are these sex differences applicable clinically? The answer to both questions is decidedly yes, of course. But we still have a long way to go. We have much to learn from the study of females, making use of the lifelong changes in their reproductive conditions as experimental variables. We also have much to learn from animals, especially if we apply what we know about their social lives. However, the challenge in all of these studies is not first to look for some mythical neurological entity called pain experience and then to learn how sex modulates it, but rather to seek to understand the rules by which sex influences all of biology's mutually modulatory factors – social, psychological, physiological, cellular, molecular, and genetic – that collectively create the motivating circumstances we designate as pain. It appears almost beyond doubt that on the one hand these factors interact to make women more vulnerable to these circumstances than men, but on the other hand that women have more varied mechanisms for balance. Happily, the details of these sex differences at all levels biological (social to genetic) are now emerging in a rapidly growing body of literature that promises new insights into and applications for the individual person, male or female, in persistent pain.

In this response, I discuss the roles of the peripheral afferent drive in the maintenance of persistent pain, the concept of pre-emptive analgesia and the importance of the brain, the detailed involvement of which in pain is far less well understood compared to the events in the spinal cord. A comparison of pain to other sensory modalities is then made together with a discussion of learning and pain. These facets of pain are discussed in the light of treatment strategies for this condition.

The commentaries on our target article have raised important issues about central neuroplasticity and its role in persistent pain states. Some suggest that central neuroplasticity plays nothing more than a minor role in persistent pain, while others argue that persistent pain depends critically on peripheral inputs for its maintenance. Some stress that persistent pain relies to a large extent on changes in the brain and on centrifugal inputs from brain to spinal cord, whereas others argue that it depends on alterations in inhibitory as well as excitatory systems. We attempt to address each of the commentators' points, while defending our position that central neuroplasticity is critical to pathological persistent pain states.

The theme of my target article was dysfunction of inhibition in the spinal cord as an important factor in the development of chronic pain states. Some commentaries focused on the role of more central mechanisms and the limited usefulness of animal models for understanding mechanisms of human pain. More specific comments concerned the roles of GABA and cholecystokinin in pain control.

Certain patients with a possible contribution of the sympathetic system to pain may not fit the definition of complex regional pain syndromes (CRPS), which raises the question of terminology for those patients. To further clarify the relationship between the sympathetic system and pain, apart from the need for placebo studies, there remains an urgent need for a satisfactory definition of the criteria for a complete sympathetic block. It also remains uncertain whether a change in the discharge pattern of sympathetic fibres underlies the changes in sympathetic organ function, often found in patients with CRPS.

The locality assumption (LA) seems rather awkward, especially when one considers centres of neuronal specialisation as defined by observed CNS activity. It is clear from electrophysiology that extra-striate functional compartmentalisation (modularity) is rather less well-defined than first thought; neuropsychological assessment attaching significance to varieties of preserved behaviour also reveals that some basic flaws must be inherent in current reasoning supporting LA.

There is a distinction between locality and modularity. These two terms have often been used interchangeably in the target article and commentary. Using this distinction we argue in favor of a “weak” modularity. In addition we also argue that both PDP-type networks and box-and-arrow models have their own strengths and pitfalls.

Although Farah makes a convincing case against the tenability of the locality assumption, she does not propose alternative research strategies that do not rest on this assumption. It is proposed here that we may profitably exploit individual differences in neuroanatomy and behavior. In combination with the use of adequate genetic methods, this approach – termed microphrenology – does not need a locality assumption.

This commentary seeks to place Farah's (1994) arguments in the historical context of ideas about mind-brain relationships. It further seeks to draw a conceptual parallel between the issues considered by Farah in her target article and questions which have concerned neuroscientists since the nineteenth century regarding the functional organization of the brain. Specific reference is made to the relationship between use of the concept of “locality” in cognitive neuropsychology and use of the concept of “localization” in neuroscience.

We respond to Farah (1994) by making some general remarks about information encapsulation and locality and asking how these are violated in her computational models. Our point is not that we disagree, but rather that Farah's treatment of the issues is not sufficiently rigorous to allow an evaluation of her claims.

Farah argues that nonlocal models explain clinical data better. However, the locality assumption does not seem so implausible if different sorts of data are taken into account. In particular, priming experiments using evoked response potentials support modularity. I describe some ERP studies relevant to this issue.

The central claim of my original target article was a modest one (that modularity does not always hold) but it was misinterpreted as a much stronger one (that modularity never holds). Further confusions arose from multiple valid usages of the term “modularity” and the similarity of the terms “locality” and “localization.” Despite the limited nature of the claim, I maintain that it poses a stubborn problem for neuropsychology, not to be dispelled by new empirical methods or a priori reasoning.

When Genotype × Environment (G × E) interactions are present, heritability estimates are not interpretable. Mealey cites abundant evidence for G × E interactions in the etiology of sociopathy, thereby completely undermining estimates of the heritability of sociopathy which form the foundation of her model. Without proper evidence for a genetic basis of sociopathy, Mealey's sociobiological model collapses under its own great weight.

Mealey's (1995a) psychological explanation of the sociopath's antisocial activity appeals to an incomplete or nonstandard theory of mind. This is not the only possible mechanism of mental state attribution. The simulation theory of mental state ascription offers a better hope of explaining the diverse elements of sociopathy reported by Mealey.

Mealey's article on the sociobiology of sociopathy raises questions about the relationship between adaptiveness and abnormality. Although sociopathy may be adaptive, simply defining medical and psychiatric disorders in terms of evolutionary function is problematic. Rather, sociopathy may be characterized as a relatively atypical disorder, an entity to which it may be useful to extend the metaphors of the medical model.