Last updated: Sat, Feb 22, 2025
[The literature on CNS circuits involved with pain sometimes refer to "affective," "emotional," and other such constructs. Some brain structure is characterized as "affective," etc., and so activation of this structure implies a link to "affective" processes. Although such characterization is based on a body of varied research, nevertheless the meaning is based on operational definitions (of "affective," etc.) as made operational in the research. E.g., unpleasant smells are deemed to evoke the "affective dimension" of pain. This interpretation is a stretch in the context of the individual research result, and the characterization of a CNS locus as an "affective center" is based on the melange of operationalizations. In particular, it is probably dangerous to think of "affect" or "emotion" in this context in the same way that one thinks of it in vernacular use of the terms. In addition, much of the basic research on which these inferences are made is on "animal models," where only external observations are possible. Also, the concept of "dimensions" of pain experience is inferential, apparently based on such evidence as the statistical analysis of response sets to instruments such as the McGill Pain Questionnaire.]The Pain Matrix is based on the chapter by Apkarian, A. V., Bushnell, M. C., and Schweinhardt, P., "Representation of Pain in the Brain", which is Chapter 7 in McMahon, S. B., Koltzenberg, M., Tracy, I., and Turk, D. C., "Wall and Melzack's Textbook of Pain", Elsevier Saunders, 2013. I take that chapter to be not far from a mainstream view within the community of pain science, and the current critique herein applies not only to that chapter, but also to many other writings about the emotional or psychological "dimensions" of pain.
My first criticism concerns some very fuzzy ideas contained in the Apkarian et al. analysis. The fuzziness involves not the sensory aspects of pain (location, intensity, duration), but the non-sensory aspects. The authors use the terms "psychological," "emotional," and "cognitive," for example, to distinguish non-sensory aspects of pain cognition, but they never define what they have in mind when they use these terms.
When, for example, they speak of "psychological" aspects1, they are referring specifically to a group of experiments that tested the effect on perceived pain intensity of directing the subjects' attention to or from the painful stimulus. This raises questions about the use of the term "psychological." If they are testing attention and perceived intensity, why are these subsumed under the term "psychological?" Does that category include other cognitive processes, and if so, what do they claim about the larger category? In other words, why call it "psychological," a word that supports a very wide range of interpretations? Why does "psychological" have a role in understanding activation of neurons in different structures of the brain?
Similarly with the category "emotional." When they use this term in a context that provides any hint as to what they mean by "emotional," they are referring to another group of experiments that tested the effect on perceived pain intensity of such manipulations as "looking at emotional faces, listening to unpleasant music, or smelling unpleasant odors...."2 Again, if they are testing what we might call "induced mood," why is this called "emotion." Certainly emotion in pain is much more than externally-induced mood.
Similarly again with "affective." When they use that term they seem, where anything can be inferred, to mean the pleasantness or unpleasantness that is perceived.
Taken together, their repeated use of these fuzzy categories (with no definitions) suggests either that their analysis of the pain matrix is embedded in a set of unspecified beliefs about these categories, or that they have not scientifically come to grips with these categories.
In Studying the Brain, I've summarized the technologies that are used in physiological analyses of the brain. Each of these technologies reveals certain information about the brain but conceals other information. In particular, brain imaging techniques are able to reveal the entire volume of the brain, but are not capable of revealing brain activity that occurs quickly or that involves small brain features or features that are small and close together. Similarly EEG and MEG are able to reveal very fast activities, but are not able to penetrate to the center of the brain (where much limbic/emotional function occurs), nor are they able to detect events in structures that are small and tightly packed.
Although the Apkarian et al. analysis does at least suggest these blind spots in the technology3, the blind spots have a great significance in limiting the extent of our understanding. This seems to me to be a topic worth addressing explicitly. My section A Rational Model of Emotion and Pain attempts to address this shortcoming by looking to affective neuroscience.