Rest: It’s not what you think it is
(The basics of the argument)
A common inference in the folkways of psychology is that a passive or resting state, wherein one simply ‘times out’ from activity and sits, is a precursor to sleep, and would likely be marked by brain states that signify a transition to unconsciousness. However, recent neurological findings have demonstrated that so called resting states are not related to sleep, but actually involve the activation of multiple neural structures that in concert accentuate the retrieval and processing of semantic information, or in other words, rational and alert thinking. Notably, this ‘default mode’ of neural activity not only correlates with the heightened retrieval and processing of information, but also correlates with the somatic state of generalized muscular relaxation.
But these observations create a paradox. If common workaday behaviors involve and require a heightened access to general knowledge, or in other words a demand to think rationally, why do similar neural and somatic states not concurrently occur? Conversely, if muscular tension or stress occurs because of conflicts entailed by a need to evaluate information or semantic knowledge (e.g. which fork on the road to take), why is tension notably absent when semantic reasoning is markedly accentuated (and presumably entails similar conflicts) when one is passively sitting with eyes open or closed? What is the common element separating such behavior that causes such a striking demarcation between states of brain and body?
The solution I propose is that semantic knowledge alone is not primarily responsible for tension, but requires the added appraisal of a contrasting source of knowledge that is generally not accessed during resting. In other words, tension is not primarily due to a conscious or non-conscious appraisal of semantic knowledge alone, but requires the combined and contrasting appraisal of abstract properties of information that gain value through neural processes that do not sub-serve language. A definitive answer to these questions require a systematic neuro-biological theory of rest that integrates the activity of the central and peripheral nervous systems and provides hypotheses that may be tested empirically. Below is my take on how such a theory will eventually evolve.
A Bio-psychological theory of resting
Recent research on the neuro-psychology of learning has revealed that incentive motivation is comprised of two neurological systems inter-operating concurrently that are distinctive psychologically and physiologically. The first system represents procedural or semantic based knowledge, and corresponds roughly to the means end expectancies that comprise rational knowledge. The second system represents knowledge of means ends discrepancies, or novel unpredicted events that have salience to an individual (Berridge, 2001). These systems operate differently, and can be dissociated not only theoretically, but also practically. The question is; what are the global physiological correlates of such systems when practically dissociated? Though no experiment has yet been performed to directly answer this question, one can make strong inferences from recent experiments that have in practice though not in design observed this dissociation.
To demonstrate this, consider a simple mind experiment. Imagine an individual beginning his or her working day in a separate office. The individual passively views a relatively static visual field (computer, phone, note pad, pen, pictures etc.) and has an established and fully predictable schedule of work. The individual also is aware of an entirely different set of choices that represent activities that have on going outcomes that are wholly unpredictable. These may represent checking email, conversing with coworkers, reading the paper, making phone calls to friends, etc. This represents a typical working day that includes a myriad decisions, but none particularly ‘threatening’. Nonetheless, it would be expected that an individual would find this type of daily agenda stressful. Now, if the individual ‘timed out’ and merely stared ahead and rested quietly, the neurological correlates to this ‘resting state’ would not represent deactivation patterns representative of sleep, but unique activation patterns of multiple areas of the brain that in sum accentuate semantic or procedural thinking (Raichle, 2001; Greicius, 2003). In other words, to take a time out and rest does not approximate sleep, but rather facilitates alert and rational thinking. Moreover, it is well demonstrated that the somatic or neuro-muscular correlates to passive resting states encompass generalized muscular relaxation (Holmes, 1984, 1988). In combination with its neural counterpart, this may be defined as a global default ‘resting’ state that encompasses brain and body.
The fact that the effective appraisal of semantic memories or knowledge does not interfere with but is rather associated with muscular relaxation raises the question as to why muscular tension or ‘stress’ occurs in workaday environments that are full of demands to think rationally, and are not nominally stressful. The answer may be quite simple, and equally as testable. An individual who is sitting alone and resting will not be affectively primed (i.e. get aroused) to prospectively engage in events that have a discrepant value (checking email, etc.), since any likelihood to imminently engage in them is effectively excluded. Thus he or she has no need to choose between events that have discrepant value or rational value. In other words, the only difference between an individual at rest and an individual at ‘work’ is that the latter must continually choose between options that represent different sources of incentive value: an ‘affective’ source of value (represented by the priming activity of a dopaminergic based motivational system) and a ‘rational’ value represented by cortically mediated procedural knowledge or means-ends expectancies. But what would muscular tension have to do with such choices?
The answer may lie in the well-accepted concept of the ‘somatic marker’. Separately developed by the neurologist Antonio Damasio and the psychologist F. J. McGuigan. This concept holds that the non-oral musculature provides information that is interpreted linguistically by the brain, and nonconsciously mediates choice by allowing one to parse between alternatives prior to their rational consideration. If alternative choices have incommensurate sources of value, then tension may ‘build’ and ultimately force choice in one direction or another. Thus, tension occurs because one literally cannot make up his or her mind. The solution and resulting procedure is obvious: to avoid tension, merely avoid the small nonconscious dilemmas that we encounter throughout our day.
Thus a worker can ‘put off’ considering all choices that are largely mediated by novelty or discrepancy to a set period of the day. This means that idle conversations, checking email, reading magazines, etc. must be performed only at certain established times (say, around lunch or at the end of a working day). The alternative to the remaining predictable fare of work is to take a time out and do nothing. By avoiding or removing ‘distractions’ or affective choices, one’s muscles will necessarily relax and the brain will revert to a ‘default’ mode that accentuates clear thinking. This procedure, which I call ‘radical’ relaxation, derives from the hypothesis that muscular tension derives from separate ideas or motives whose value stems from different neurological sources. That is, the activity of musculature occurs to expedite choices that cannot be logically made between two different sources of incentive value. Cognitive styles that combine the two will necessarily result in tension. For cognitive styles that separate them persistently, relaxation will follow and also persist.
Berridge, K. (2001). Reward Learning: Reinforcement, Incentives, and Expectations, The Psychology of Learning and Motivation, 3, Academic Press, New York
Damasio, A. (1994) Descartes Error: Emotion, Reason, and the Human Brain. Avon: New York
Grecius, M. D. Krasnow, B., Reiss, A. L., & Menon, V. (2003) Functional connectivity in the resting brain: A network analysis of the default mode hypothesis, Proc. Natl. Acad. Sci. 100, 253-258
Holmes, D. S. (1984). Meditation and somatic arousal reduction. A review of the experimental evidence. American Psychologist, 39(1), 1-10
Holmes, D. S. (1988). The influence of meditation versus rest on physiological arousal: a second evaluation. In Michael A. West (Ed.) The Psychology of Meditation, Oxford: Clarendon Press
McGuigan, F.J. (1976). The function of covert oral behavior in linguistic coding and integral information processing. In Kurt Salzinger (ed.) Psychology in progress: an interim report. Annals of the New York Academy of Sciences, 270: 57-89
Raichle, M. E., MacLeod, A. M., Snyder, A. Z., Powers W. J., Gusnard, D. A. & Shulman, G. L. (2001) A default mode of brain function, Proc. Natl. Acad. Sci., 98, 676-682
Relaxation and Muscular Tension: A Bio-behavioristic Explanation
A. J. Marr
ABSTRACT: Relaxation and muscular tension are commonly ascribed to cohere to metaphorical principles of stimulus-response that presumably are isomorphic with neurally based reflexive mechanisms that are not modifiable by learning. However, purely S-R principles have been progressively replaced in modern learning theory with expectancy or discrepancy models of learning that do not recognize separate neural processes that subserve operant (R-S) and respondent (S-R) conditioning. An alternative explanation for the relaxation response and muscular tension is provided that is derived from these principles, and demonstrates theoretically and through practical procedure that muscular tension is universally influenced by simple bio-psychological processes of incentive motivation or learning.
Key Words: Coverant, relaxation response, Pavlovian incentive motivation, affect
Introduction
In contemporary psychology, muscle relaxation has commonly been interpreted as the end product of reflexive mechanisms that are activated by obscurely defined mechanisms of attention. These interpretations have an ancient lineage, and their psychological representation as ‘meditation’ has had a long and exhaustive history as a topic for empirical investigation in psychology. The fact that selective focusing on simple stimuli or cognitive precepts (e.g. mantras) correlates with relaxation is undisputed, but has been historically confounded by claims of additional mental benefits (e.g. unique states of consciousness) that supercede the bounds of conservative empirical inquiry. Extensive surveys of the critical literature on meditation (Holmes, 1984, 1988) have demonstrated that the dependent variable for all meditative protocols is no more than generalized muscular relaxation. However, the correlation between focused attention and relaxation has continued to imply an underlying causal relationship. Nonetheless, the neuro-psychological connection between focused attention and relaxation has never been empirically demonstrated, and has been explained instead by wholly inferred neurological linkages between discrete attentive states and relaxation that are equivalent to Pavlovian stimulus-response (or S-R) bonds. That is, a ‘relaxation response’ (Benson, 1974) occurs as a reflexive response to concentrated or focused attention, and is the physiological opposite to the ‘flight or fight’ response that is embodied as anxiety or stress.
It is the position of this article that S-R explanatory mechanisms for the somatic effects of the relaxation response and its opposite of muscular tension are false, and that relaxation and tension are not dependent upon discrete and consciously perceived Pavlovian-like stimuli of ‘attention’, but rather upon indiscrete and nonconsciously perceived patterns of information or ‘expectancies’ that cohere with modern neuro-psychological models of learning. This new model for the relaxation response and tension is easily testable and refutable, and suggests strong procedures for behavioral or self-control.
The Illusion of Stimulus-Response
Relaxation as well as muscular tension follow the theoretical principle, long established in psychology, that non-consciously elicited somatic responses follow and are solely determined by initiating or stimulus events. Indeed, until the last twenty years, the common view was that learning was bifurcated into behaviors that were initiated by stimulus or antecedent events alone (the stimulus-response or S-R model) or were initiated by response events that had a subsequent stimulus outcome (the response-stimulus of R-S model). These disparate principles produced the respective data languages and procedures of classical (Pavlovian) and operant (Skinnerian) conditioning. The behavioral metaphors of classical and operant conditioning in turn extended to a commonly accepted physiological metaphor that localized S-R and R-S processes in separate parts or organelles of the brain. Thus, S-R or Pavlovian processes were localized in midbrain structures that operated in an analogical manner, and R-S or Skinnerian processes were localized in cortical processes that operated in a logical or computational manner. These processes in turn acted independently of each other. Finally, S-R and R-S metaphors provided a taxonomy wherein certain modalities of behavior could be classified. For example, operant behaviors encompassed the conscious control of the striated musculature, whereas respondent behaviors reflected the nonconscious control of the smooth musculature.
The problem however with this metaphorical scheme is that it is wrong on all levels. In modern learning theory, classical and operant conditioning are distinguishable through their unique data languages and procedures (Staddon, 1993), and not through separate identifiable processes. Thus the behavioral metaphors of distinctive S-R and R-S processes is wrong, since such processes have never been isolated to begin with. Secondly, the physiological metaphor that canalizes classes of behavior to separate cerebral organelles is also wrong. ‘Higher’ and ‘lower’, cortical and midbrain structures are exquisitely interdependent, and behavior is informed mutually by associative learning (or thinking) and affective processes that reflect the hedonic attributes of brain and body, whether they reside in neurochemical, neuromuscular or other somatic events (Panksepp, 1998, Damasio,1994). These changes require a new perspective of how learning works.
In its rudiments, a learning theory describes how experience shapes behavior, and as a matter of principle must account for the antecedent and consequential events that shape behavior as well as all behavior from the molar to the molecular. The elimination of the metaphors that command a blinkered perspective on learning provide for new metaphors of learning and reinforcement that integrate these pictures in a unified view of learning. Present day single factor or unified theories of reinforcement or reward (Donahoe and Palmer, 1993, Berridge, 2001) are distinguished from S-R and R-S models by integrating antecedent and consequential events with the behavior they frame. Consequential events are also termed contiguous or contingent events. In particular, for every behavioral response, whether overt or covert, a resulting environmental or behavioral event is present and will influence behavior. For example, if a student is tense, he may postpone taking a test, or if an individual is tense while deciding whether to buy a new car, he may make a decision quickly to relieve the tension of indecision. In both cases, the somatic behavior of tension will recur because it historically led to a contingent or ‘reinforcing’ outcome, as when an individual after rumination fixed by tension settles on or perhaps rationalizes an appropriate choice. Likewise, a lack of contingent outcomes to behavior may also influence behavior. For example, avoidance behavior is reduced if an individual has no historical opportunity for escaping in similar circumstances, as the well known phenomenon of ‘learned helplessness’ demonstrates (Seligman, 1972), or test anxiety may be reduced by a ‘time out’ procedure that withholds potentially reinforcing responses, such as escape.
Given this new perspective that postulates a universal role for contingency or contiguity in learning, or an S-R-S model, all behavioral aspects of brain and body from molar to molecular become subject to a new unifying metaphorical vision. But, in the case of the ubiquitous somatic states of muscular tension and relaxation, it also requires a clarification of what that behavior is.
The Non-Response of Relaxation
The first empirical appraisal of muscular relaxation and tension dates from the work of the psychologist Edmund Jacobson (1938) and his colleague and student F. J. McGuigan (1976,1997), who postulated that muscular tension was a behavioral response that nonverbally modulated discriminative or contingent outcomes, namely patterns of thinking and the action patterns they entailed. In turn, relaxation did not modulate such outcomes, and thus was not a behavioral response. Rather, relaxation denoted the relative inactivity of the striated musculature. That is, relaxation merely entails that the musculature, which is comprised of more than a thousand groups of muscular fibers, is for the most part doing nothing. Moreover, such inactivity is unmediated by other cognitive or proprioceptive stimuli, as in the conscious and nonconscious events that permit one to get relaxed and stay relaxed.
The Jacobson-McGuigan hypothesis for relaxation and tension has never been directly challenged, but nonetheless was overshadowed by biological models of emotion (Cannon, 1927) and stress (Selye,1956) which incorporated tension as one of a constellation of physiological events that were not at root behavioral, but represented instinctive or reflexive responses to an environmental ‘demand’. A similar S-R model was later provided by Benson (1974), who postulated that relaxation was a reflexive response to attentional focusing, and since it was not influenced by contingency, was also not behavioral.
The problem however with S-R linkages of demand-tension and attention-relaxation is that the independent variables of demand and attention have mutually exclusive definitions. Thus ‘demand’ may represent a surprising or threatening event that elicits muscular tension as a priming response leading to ‘flight’ or ‘fight’, or demand may represent non-surprising and non-threatening choices that will elicit tension not to prime flight or fight, but to differentially influence thinking and thus choice, as the Jacobson-McGuigan model would predict. However, for common explanations of stress, muscular tension in all circumstances is subsumed under the fight or flight response, thus marginalizing the Jacobson-McGuigan model as a complimentary definition of demand, and leading to the somewhat awkward conclusion that the tension elicited while choosing between the myriad unthreatening choices of a workaday existence reflect a mere species of fear or anger, or of flight or fight.
A more complicated issue in the case of the relaxation response is whether attention can be defined as an indivisible stimulus event. The critical issue underlying the S-R mechanisms that are implied by meditative protocols is whether selective or focused attention is a necessary independent variable that co-varies with a dependent state of relaxation, and whether attention should actually be used in the first place as an independent variable. In his survey of the meditation literature Holmes (1984, 1988) concluded that it does not, and that relaxation can occur just as surely and effectively by simply avoiding thought while sitting in a quiet place, or in other words, resting. Moreover, ‘attention’ is not an indivisible process nor can it be used as such, but is rather a taxonomy for the distinctive conscious and nonconscious neural processes that underlie conscious choice (Kruschke, 2001) that represent distinctive independent variables in themselves. Specifically, attention may represent semantic or declarative thought that may be complex (as in paying attention to a move in chess) or attenuated (as in repeating a word or phrase), or it may represent the non-conscious gating of information or orientation due to stimuli that represent abstract properties of information (as when attention is directed to a novel or ambiguous event, or distraction). The bifurcation of attention into two independent variables along with thought avoidance (rest) leads to the observation that the relaxation response hypothesis encompasses not one but three different independent variables that correlate with muscular relaxation: a repetitive mental or physical activity, passively ignoring thoughts, and avoiding distractions. These variables of course are of course embedded in the formal definition of Benson’s relaxation response hypothesis (1974) yet a characteristic of research on the relaxation response or meditation is that the variables of ‘ignoring thoughts’ and ‘avoiding distractions’ have not been adequately controlled or defined (Holmes, 1984, 1988) since relaxation is wholly attributed to repetitive mental activity.
To illustrate, The primary variable of ‘focused attention’ postulates that relaxation will follow perseveration in attenuated declarative thought (i.e. repeating a nonsense word) in conjunction with reducing thoughts, yet the second variable, namely avoiding semantic reasoning or thought (i.e. clearing your mind of thought), also correlates with relaxation, but has been demonstrated to do so independently of the need for repeating a word or phrase. In other words, a situation in which thoughts are avoided but attenuated thought is not required, will be no less effective than a resting state conjoined with such thinking (Holmes, 1984, 1988). Thus, the relaxation response must require thought avoidance to occur conjunctively with attentional focusing (attenuated thinking), but as Jacobson and McGuigan demonstrated, thought avoidance does not require focused attention to result in relaxation! Compounding this difficulty is that ‘distractions’ are not at root cognitive states but ‘affective’ states, and because they are conflated with cognitive behaviors are not considered to represent a class of variables that should be separately controlled. That is, an individual who is resting or meditating may clear his mind of thoughts, but nonetheless be aware of the need to avoid distracting events in order to be and stay relaxed. Finally, it may be argued that even the much vaunted correlation between attenuated thought and relaxation does not exist. In particular, individuals in a resting or relaxed state demonstrate an increased capability (Grecius 2003, Raichle, 2001) to simultaneously perform complex reasoning. In other words, effective thinking or reasoning is not only compatible with relaxation, but is dependent upon relaxation. If this latter observation is true, it would invalidate not only the relaxation response hypothesis, but also McGuigan’s position that muscular tension is necessary for complex thinking. As will be later postulated, this will require a revision of McGuigan’s hypothesis..
The conclusion follows that attenuated declarative reasoning (e.g. repeating a nonsense word), because it cannot be shown to independently correlate with the initiation and maintenance of relaxation when complex thoughts or distractions are controlled, is of no more value as an explanatory variable than the element of phlogiston, an imaginary ingredient that enabled fire. Since relaxation does not require the presence of distinctive eliciting or contingent stimuli, it cannot be considered a response or a behavior, but is rather a steady homeostatic state representing the non-activity of the musculature. Combined with the empirical status of relaxation as a non-behavioral event, the explanatory framework for meditation and the relaxation response does not cohere with neurally based models of attention and learning as presently conceived. But if learning or experience is the key to understanding relaxation and tension, how can a neurally informed theory of learning recast the theory and practice of muscular tension and its control?
Parsing Reward
In modern bio-behavioral models of learning (Donahoe and Palmer, 1993, Berridge, 2001), contiguity and contingency influence nearly all behavior, including the covert responses commonly subsumed under the rubric ‘emotion’, and are presently incorporated in the data languages of behaviorism, whether Pavlovian, Skinnerian, or the molar expectancy theories derived from Tolman. The problem and challenge for theorists on covert behavioral states that include muscular tension such as fear, anxiety, or its more encompassing term of ‘stress’ is that contiguous or contingent factors are not well integrated into models of stress or more commonly are ignored entirely. This can be largely attributed to the fact that covert emotional states that include muscular tension are measured by self reports (Wilhem and Roth, 2001) that cannot be dissociated into separate somatic components, thus making the causes of these components easily conflated, and the contiguous and contingent events that follow them rendered difficult to determine. This has resulted in the incorrect parsing of all somatic components of the stress response from muscular contractions to glandular activity to represent simple S-R linkages that ignore the effects of contiguity and contingency that are now implicit even in the ostensibly S-R data language of Pavlovian conditioning. However, although learning principles may be unified to include antecedent and consequential events, incentive motivational principles are not. In other words, although behavior may occur through unitary principles that incorporate past and future events, the antecedent and consequential events that drive behavior are themselves bifurcated into very separate processes.
In the folkways of common sense, reward is something that is appraised logically, and pleasurable or painful events are pivot points that fit in an overall rational calculus of utility. Thus, we appraise ‘goods’ because of logical measures of their utilitarian and hedonic value. Moreover, our often pleasurable anticipation of attaining goods that have hedonic and/or utilitarian value, or our ‘wants’, are an inextricable aspect of the things we desire, or ‘like’. Because of this common sense or heuristic principle, behavior is a function of how we manipulate the contents or inventory of the things we desire. Thus our workaday strategies (as well as the psychotherapeutic strategies that we turn to when we fail) revolve about an economic calculus of value where motivation and emotion are relative to real and metaphorical objects of desire. By managing the content of cognition, namely what we want and how we will get it, we manage our behavior and implicitly, our emotions. The problem is that we can never get all we want, or get what we want in the time and fashion we choose. This of course is the basis of stress, a confluence of choices or demands that we are continually juggling, and more often than not failing to manage, as we proverbially drop the ball at home and at work. Thus poor decision making, and hence stress becomes an endemic and seemingly inescapable aspect of our lives. But what if mis-managing the objects of desire is not the primary cause of muscular tension, but rather the style of desire, or how we anticipate the objects we value? But to know how we anticipate, we must first define what anticipation is.
Anticipation can be appraised as a singular motivational or incentive event elicited by an object of desire that merges expectation with arousal, but according to modern neuroscience, cognitive expectations are different behaviorally and neurologically from the affective salience or felt state of arousal that is present when we are anticipating an object of value (Balleine & Dickinson, 1998, Dickinson, Smith, & Mirenowicz, 2000). Indeed, affective or incentive salience has nothing to do with the normative properties of an object of desire, but rather to abstract properties of information that are mediated by the relationship of a performance and the object itself. This information is enfolded in ‘discrepancy’ theories of reward that identify reinforcement or reward with our perception of positive discrepancies (Donahoe and Palmer,1993; Schultz, 1998), positive events or turns of events that are unexpected. More formally called Pavlovian incentive motivation (Berridge, 2001), decision making is influenced by knowledge derived natively or instinctively from abstract properties of information as represented by its relative predictability. This knowledge is represented non-verbally by the affective or hedonic value that derives from the release of neurochemicals or neuromodulators (dopamine) that modulate global areas of the brain. This process mediates the predictive expectation of a reward signaled by a stimulus, and teaches new predictions by signaling deviations from the expected (Montague et al. 1996, Schultz, 1998). But like cognitive expectancies, dopamine systems also instigate behavior. Neurologically, dopamine systems (DA) interact with the prefrontal cortex (PFC) and induce motor (i.e. behavioral) responses that suppress the DA system, reset the PFC, and return activity to baseline levels (Cohen, Braver, & Brown, 2002). Restated in behavioral terms, human beings will respond or orient to surprises or discrepancies in their environment that represent variances from what they have come to predict, and acquire this interest not through logical processes of thought, but through nativistic analogical processes rooted in the structure of the human brain.
Ultimately, anticipation, or more specifically incentive motivation is a product of separate but interactive processes that are different behaviorally and neurologically (Berridge, 2001). But this observation leads to a conundrum. If incentive or value is also rooted to abstract and analog principles that are psychologically and biologically separate from the logical objects of desire, they must invariably integrate or clash with the concrete and logical principles that we commonly believe informs value. Indeed, common sense heuristics tells us as much. Affective events cohere with rational events when we pursue an interesting project at work , enjoy a healthy meal, or pursue an artistic endeavor that will achieve the applause of our peers. But affective events may also be incoherent with rational events. Choosing between raiding the refrigerator and staying with a diet, stopping in a hallway to chat or going about one’s business at work, or watching TV rather than doing household chores all represent choices between rational and ‘affective’ choices, providing us with choices that derive from different sources of value that cannot be logically compared. In common sense terms, a day filled with such choices is a day full of ‘distractions’, endless choices that we cannot resolve logically, but perhaps do so somatically through muscular tension. But the question is, can muscular tension act covertly to force decision where logic cannot?
Somatic Markers
For cognitive associative learning, or thinking, decision-making derives from procedural knowledge that represents act-outcome expectancies. This knowledge may be represented verbally or consciously, but it may also be latent or nonconscious, and may be mediated non-verbally by the non-oral musculature (McGuigan, 1978). Muscular tension has been hypothesized to provide a nonconscious ‘allographic code’ (McGuigan, 1976) that is linguistically interpreted in the brain, and conforms to associative learning principles. Thus, we literally albeit nonconsciously ‘think’ with our musculature, a notion that has achieved common currency in contemporary theories of emotion (Damasio, 1994). Applying this model to muscular tension, it follows therefore that relaxation will occur when conscious or nonconscious information is not available that would otherwise be embodied by muscular contraction, or when such conscious or nonconscious information is suppressed or inhibited. For example, a student would sit still if there were no distractions that would otherwise cause him to move, or he may sit still because a teacher’s threat of punishment may cause him to ignore or suppress such distractions. Similarly, relaxation may occur through the avoidance of distractions (as in taking a vacation to a remote beach resort), or it may occur through the alteration or suppression of distractions (as in ignoring, reframing, or suppressing distracting thoughts and perceptions).
Associative learning is a critical explanatory component for the widely accepted notion of the 'somatic marker' (Damasio, 1994). A somatic marker is an automatic somatic signal influenced by the laws of learning (or more specifically, neocortical processes) that enables an individual to parse between alternatives prior to their rational or conscious consideration. As such, it centers the modulation of somatic or affective states to the activity of the somatosensory cortices which map the body. Somatic markers encompass affective events such as hormonal (hunger, thirst) or neuromodulator (excitement, depression) activity, or the activity of the viscera (gut feelings) or musculature (muscular tension). For example, desire may be mediated by semantic or act-outcome expectancies (i.e. rational behavior), or it may be accentuated or ‘marked’ somatically by the prior association of such expectancies with an ‘affective’ source of value, such as feeling of arousal marked by dopamine release or a literal ‘gut’ feeling. Thus, a new car is valued because it has functional properties, but also because of it’s novel appearance, sense of comfort, anticipated positive reactions of peers, etc., that represent affective events.
Somatic markers represent different somatic activation states that as recent research suggests are specific to different types of choices. For example, for semantic or procedural reasoning, the gating of attention through the release of the neuromodulator dopamine provides incentive or affective value to the appraisal of the myriad choices that in critical thinking implicitly involves novel or unpredictable outcomes. and is core to neurobiological definitions of reinforcement (Donahoe and Palmer 1993, Schultz 1998). However, semantic reasoning is markedly accentuated during states of rest where the musculature is in an inactive or relaxed state (Greicius 2003, Raichle, 2001), and peripheral feedback in sensory and motor function is not required for complex reasoning and behavior (Taub, Ellman, & Berman, 1966, Teuber, 1972). Thus, for an individual pondering a problem in science, a move in a chess game, or what item to choose from a restaurant menu, decision making is more effectively mediated neurologically without the need to include the afferent input from the musculature. In contrast, for forced choice reasoning where decision has to be made rapidly between alternatives that cannot by time or circumstance be logically compared, effectiveness in decision making is a function of the speed wherein decisions are made, hence muscular tension or ‘stress’ may occur to force decision prior to its effective appraisal. In particular, muscular tension as a somatic marker may develop through learning, and be construed to be functionally similar to a covert operant, or ‘coverant’ (Homme, 1965) in that it indirectly operates on behavior by directing choice, and is governed by nonconscious information. According to this model, muscular tension occurs as a somatic marker that expedites decisions between two choices prior to their conscious consideration. Thus, tension occurs because it is critical to rapid decision-making between rational alternatives. The question is, can tension also occur to mediate decisions between alternatives that are choices that are rational and affective? As presently conceived the somatic marker hypothesis cannot provide an answer. This is because by situating the origin of experienced somatic or affective states in associational learning processes centered in the neocortex, Damasio ignored the many sub-cortical sources of emotional feelings or affective value (Panksepp, 2003), and how they, as in the case of Pavlovian incentive motivation, may indirectly influence behavior, and the elicitation of the somatic marker of muscular tension. Thus, there is no way to test this hypothesis within Damasio’s theory since affective processes are outside its theoretical range.
Associative learning and Pavlovian incentive motivation represent entirely different yet continually present neurological sources of value that result in ends that may be additive or be at cross-purposes. For example, the act-outcome expectancies that fill our workaday lives generally entail predictable relationships between means (work) and ends (a paycheck). But the relative predictability of work may be interspersed with unexpected or ‘discrepant’ events that may cohere with and enhance the aims of work (e.g. praise from a boss, completing a work task more quickly than expected) or be incoherent (e.g. checking email, reading a newspaper, gossiping) with doing a good job. Moreover, an individual may be stimulated or ‘primed’ by the mere modeling of such events, such as expecting an imminent raise or coffee break. In other words, the cognitive representation of positive future events will also stimulate by virtue of the continuing virtual uncertainty of those events. When conscious or nonconscious procedural decision rules governing alternative means-ends expectancies (i.e. applying logic or reason to make a decision or resolve a problem) are not known or are partially known for in situations that require quick choice, then in keeping with a somatic marker hypothesis, muscular tension may occur to drive decision in one direction or another. Thus we may become ‘on edge’ when making critical life decisions. However, another and more prevalent source of tension or stress occurs when one must choose between two different and incommensurable neural sources of value. In other words, the priming of behavior (pleasurable alertness) as measured in the affective or hedonic value of certain choices may co-vary with the procedural or cognitive rules for decision-making. In our daily lives, priming occurs constantly to influence choices that co-vary with the procedural rules of behavior. That is, we know that diverting or discrepant behaviors are available at anytime. Thus we are often diverted or ‘tempted’ to engage in idle conversation, check email, read the newspaper, etc. because of the novelty or discrepancy these events entail. The continuous ‘choice-choice’ between decisions mediated by different incentive factors signals muscular tension that ultimately expedites decision. However, if an individual chooses an end that has affective or discrepant value, or in other words, ‘surrenders to temptation’, then that temptation will recur as a continuous prime for future behavior. So, by taking a time out to converse about the weather or sports, the option to do such diversions remains present throughout the day, and will continue to arouse attention, divert behavior, and elicit other somatic events (i.e. muscular tension) that will expedite that diversion. In other words, by surrendering to temptation, the temptation remains constant, and so do the discomforting somatic events that we call muscular tension or ‘stress’. So ironically, by taking time out in a stressful day to perform some diversion, we actually accentuate tension.
Finally, the postulation of tension as a covert operant that occurs and is reinforced within the discriminative or stimulus control of a specific type of choice resolves the ambiguity associated with findings of McGuigan and Damasio that implicate the musculature in decision making, and subsequent findings (Taub, Ellman, and Berman, 1966, Teuber, 1972) that do not. Rather, tension influences not decision making per se, but rather decisions that cannot be expeditiously resolved through declarative reasoning. In other words, tension does not indirectly resolve decisions, but dilemmas.
The Pragmatic Science of Stress
In sum, contemporary neuro-psychological models of incentive motivation posit separate neurological sources of incentive value, whereas the somatic marker of muscular tension occurs through associative learning to expedite choices between sources of incentive value. Following this logic, the somatic marker of muscular tension occurs not because of a discrete stimulus event or ‘S’, but as a interoceptive, psychosomatic, and operant behavior that helps to mediate decisions between different stimulus event-stimulus outcome expectancies (S-S*), or in a molecular behavioral language, response contingencies. These S-S* expectancies or response contingencies denote the weighing of separate but exclusive means-end relationships (e.g. which fork on the road to take), or they may denote different sources of neurological value. Thus, by deciding between the road to God (logical goals) or mammon (temptation), one must weigh different ‘logical’ and ‘affective’ sources of value that at root cannot be logically compared.
The fact that human incentives do not have a unitary neurological source is ironically not contrary to the common sense heuristics that humans have always used to explain their behavior as well as control stress. Indeed, to master stress is to master the skills that enable one to predict with much greater reliability the proper course of action. Thus, one becomes less nervous about driving cars, dating, or doing income taxes when the skill sets are mastered that enable one to predict outcomes like a good night kiss or avoiding an IRS audit. Humans are also well aware of the unreasoning power of ‘affective’ events, and reduce the ‘priming’ or temptation to perform them by circumscribing in time the events that they can or cannot do. Thus we set aside certain times where we can eat, gossip, watch TV, etc., knowing full well that tension would surely occur if we were continually primed to veer from our present more constructive behavior.
If we take this latter behavior to an extreme, when nearly all priming variables or ‘primes’ are sharply reduced, we have meditation. Thus, by avoiding all conflicting cognitive alternatives, a resting state or state of inactivity follows because we are not deciding between those alternatives (i.e. thinking) with our minds or with our musculature. By crowding out other conflicting cognitive activity, repeating a nonsense word co-opts the need to think or the awareness of thought, and by not thinking, relaxation necessarily follows.
The Conflation of Affect and Attention
Although relaxation is the main dependent measure for focused attention, relaxation is generally felt as an affective or pleasurable state, and affective states such as reported ecstasy or pleasure are also commonly assumed (Csikszentmihalyi, 1990) to be inherent in or caused by attention itself. Indeed, the release of endogenous opiods or endorphins are correlated with resting or relaxed states (Infante, 1998, Maclean, 1997), thus likely accounting for the common attribution or felt presence of a hedonic state to the inactivity of the musculature. However, the stimulus-response model that attributes affective states to focused attention is incorrect because it conflates the separate neural processes governing affect and attentional focus. If focus is conservatively defined through the modulatory or activating processes for global states of the brain, then rehearsing a nonsense word or mantra will cause less neural activity than rehearsing say a role from 'Hamlet'. However the degree of cortical activation does not in any way co-vary with affective experience, and indeed cannot do so since affective experience is not rooted in a neo-cortex that is in fact enervated (Panksepp, 1998). Rather, it is the content or computational quality of that activation that matters, and how that content interfaces with lower or midbrain systems that do modulate affective experience. Thus one may be focusing on a stimulus or narrow range of stimuli and report a range of affective experiences from panic to ecstasy. For example, an individual running from a predator or focused on completing an income tax return on time may be acutely focused on a means of avoiding the bite a lion or the IRS, yet would feel anxious rather than relaxed. Contrariwise, an individual may be focused on creating art, scaling a mountain, or performing a delicate surgery and feel ecstatic. This final example bears witness to the fact that tasks that challenge an individual to the limits (but not surpassing those limits) of his capability, or a matching or demand to skill, will result in a high degree of prediction error that reflects the moment to moment uncertainty that a critical goal (e.g. artistic inspiration, staying on the mountain, losing the patient) will be attained or lost. Indeed, these ‘flow’ states (Csikszentmihalyi, 1990), which are also incorrectly attributed to focused attention, are actually instantiated by high levels of dopamine release derived from continuously perceived positive prediction error. This fact has been confirmed by fmri studies (brain scans) of individuals in flow producing situations such as gambling, creative pursuits, etc. (Fried et al. 2001, Koepp et al. 1998).
Finally, attentional focus is often conflated with nonconsciously perceived cognitions or expectations that in turn elicit affective states. Cognitive expectations of relief, relaxation, or other somatic or ‘psychic’ events influence response persistence, content, and its subsequent interpretation. This ‘placebo’ effect can also influence brain function on a biological level (Leuchter et al., 2002; de la Fuente-Fernández et al., 2001), with an attendant increase in affective tone that can be subsequently and incorrectly interpreted as due to attentional focusing alone. Specifically, dopamine levels may also increase as an individual is ‘primed’ to respond to anticipated positive outcomes as well as positive discrepancies (Martin-Soelch, C. & K. L. Leenders. 2001). Nonetheless, as with the perception of cognitive discrepancies, conscious or nonconscious expectations are generally not experimentally controlled in studies on the efficacy of meditative protocols.
Popular Psychology and Therapeutic Method
As these examples attest, the substitution of a neuro-psychology of affect with a psychology of metaphorical inferred processes ungrounded to the actual workings of the brain in ‘action’ has resulted in widely accepted psychological constructs (flow, meditation, etc.) that unrealistically account for human behavior. In particular, in popular psychology neuro-psychological accounts of behavior are noticeably absent, and are replaced by discrete mental objects (e.g. pride, will power, self-esteem, focused attention) or metaphors for causality that follow hydraulic, computational, or other vitalistic schemes. That this mentalistic thinking scarcely coheres to how the mind actually works has been an impediment to a true understanding of behavior (Panksepp, 2000) and as it must logically follow, the development of effective procedures that may predict and control behavior.
The explanatory principles of meditation are case in point. Meditation, whether transcendental meditation or the ‘relaxation response’ are effective because they presume causal relationships between relaxation and a mode of thinking (i.e. nonsense words, mantras) that is nearly eliminative of thinking. But this premise is false, for it is not thinking but conflicted thinking that is at the root of muscular tension. Thus the end result of meditation, namely relaxation, can occur by changing the style rather than reducing the content of cognition, and may be demonstrated through an easily replicated procedure.
Tension of course cannot be eliminated in our daily lives because we live in an uncertain world. Thus, informational sources of conflict cannot be absolutely avoided, as we must all anxiously face the uncertainties of death and taxes. However, conflicts between ‘affective’ and ‘logical’ choices can be avoided, and in our workaday worlds they are generally not because we often cannot conceive of how they may lead to stress. Indeed, taking a time out to check email, gossip, read magazines, watch TV, etc. are looked as a palliative for stress rather than a cause. Ironically, by doing the opposite, and avoiding for preset hours during the day all possibilities of engaging in events that are primarily or exclusively selected because of some novel, discrepant, or otherwise affective value, relaxation will naturally follow since a major source of conflicted thinking is eliminated. So if the alternative for doing logical acts is doing nothing at all, or in other words ‘a time out’, one has only to weigh the logical alternatives of doing something or doing nothing, choices that are far easier and less stressful to make since affective value is left out of the equation. Finally, since effecting a change in cognitive style can be made without withdrawing from workaday behavior, as would be the case with meditation, this procedure can hypothetically provide for a state of relaxation throughout a working day, and represent a far more effective method for emotional control.
Discussion
The 20th century has been labeled the ‘Age of Anxiety’, yet for the affluent classes of the modern world, the number and severity of threats to life and limb are far less than the dire events from pestilence to war that confronted our ancestors. A more persuasive case may be made that the modern age is not characterized by more pernicious threats to life and property, but rather by more pervasive choices, bad and good, which continually impact every facet of our lives. Indeed, the significant increase in modern times of depression, anxiety, and stress related disorders has been far more related to a increase in choices (Schwartz, 2003) rather than threats. In other words, modern stressors are not characterized by more threats, but more options. Perceived threats of course are different in kind from perceived choices, and so too it may be argued are the somatic events they elicit. For example, strong emotional reactions such as fear, rage, panic, etc. are the significant and separately distinctive emotional components of the 'flight or fight' response that do engage (Panksepp, 1998; Ledoux, 1997) reflexive mechanisms, as individuals must be able to immediately prime their behavior to respond to threatening circumstances. Specifically, the attribution of tension to sub-cortical processes rooted to organelles such as the amygdala links tension to novel, surprising, threatening, or ambiguous stimuli that are not characteristic to the cortically mediated choices that comprise the great majority of the stressors in our daily lives. In these latter cases, muscular tension may precede ‘fight’ (anger) or ‘flight’ (fear) responses or may not be associated with such primary affective states at all. That is, the somatic responses that may be elicited from such choice-choice options, namely muscular contraction, do not necessarily include the hormonal and neural responses that are characteristic of more primal emotions such as anger and fear. More specifically, it may be argued that choice-choice options between simple alternatives (e.g. what hat to wear to go to church) initiate response classes that are different not only in magnitude but also in type from similar options that involve much more severe outcomes for a poor choice. Further, a case may be made that muscular contraction does not represent an emotion at all, since it does not differ in kind from voluntary muscular activity (running, talking), but only in terms of its micro-behavioral characteristics that render it intractable to conscious observation. That is, just as voluntary movement of the gross musculature is not emotional in itself, the same conclusion must be applied to subtler muscular activity that is controlled as well (albeit non consciously) by the same perceptual events. Because the transition from low to high levels of stress represents changes in classes of responses as well as a gradations of similar responses (e.g. getting more and more tense), the 'flight of fight' response so commonly identified by psychologists with stress is inapplicable to the type of stress reactions (i.e. muscular tension) that is elicited by non-threatening situations. Thus, for low-level stressors, muscular tension may be presumed to be the dominant physiological response (McGuigan, 1978) and may be understood through processes of human conditioning or learning.
A learning based perspective on stress has the benefit of drastically simplifying stress control procedures by providing an explanatory model for everyday muscular tension that engages observable conditioned behavior rather than reflexive, metaphorical or otherwise inferred events. Thus muscular relaxation may be controlled by avoiding information (meditation, time outs, vacations), changing or suppressing information (reframing, positive thinking, psychotherapy, coping skills), and learning how to perceive information. The last example implicates progressive relaxation (Jacobson, 1938), which entails the progressive relaxation of different parts of the body as a means of learning how to perceive the proprioceptive stimuli that permit the discrimination and control of relaxed states.
This new taxonomy for muscular tension engages modern bio-behavioral learning theory (or less strictly, a radical behaviorism) since it defines tension and the mediating biological sources of incentive value as behavioral and not just physiological events (Barnes-Holmes, 2003; Moore, 2001). That is, the somatic responses that underscore incentive motivation (e.g. dopamine release, muscular tension) are dependent upon discriminative and contingent events that may be consciously or non-consciously perceived. Specifically, it expands Damasio’s somatic marker hypothesis by postulating that the somatic event of muscular tension occurs not just to expedite decisions between rational values that are cortically instantiated, but also to speed decision making between disparate and often conflicting axes of rational and affective values that respectively derive from cortical and sub-cortical structures.
Nonetheless, because neuro-psychologically informed models for stress, and indeed for all psychological phenomena are new, they have not been as of late persuasive or even of passing interest to the vast majority of psychologists. The reason of course is simple. It is much too easy to conflate the causes as well as components of stress, and make simple muscular tension of the same but lesser stuff as an anxiety response that may engage different causes and different or additional somatic responses. Thus a contingency based analysis of simple muscular attention is supplanted by reflexive or S-R mechanisms under which all stress responses are subsumed. If this is to change it ultimately must be through reasons pragmatic as well as theoretical or empirical. In other words, the explanatory schemes that we accept are forced upon us not only because of the simplicity of their logic and the empirical evidence that supports that logic, but also because of the effectiveness of the methods those schemes entail. Thus we must accept and understand biological theories of disease, of quantum mechanical explanations of the universe, or of the evolution of the human genome not just because the empirical facts point to them, but also because the technological features of our world depend upon them. The studied ignorance of bio-psychological accounts of human nature near universally displayed by present day psychologists is no different from that displayed by the clergy who rejected Galileo’s offer to look through his telescope at the true nature of our solar system. But this unreasoned stubbornness will disappear, not with the plaintive logic that a despairing Galileo offered to his critics, but with the unstoppable power of method that is entailed by the truth.
References:
Balleine, B. W., and Dickinson, A. (1998) Goal-directed instrumental action: Contingency and incentive learning and their cortical substrates. Neuropharmacology, 37(4), 407-419
Barnes-Holmes, D. (2003) For the radical behaviorist biological events are not biological and public events are not public, Behavior and Philosophy, 31, 145-150
Benson, H. (1974) The Relaxation Response, New York: Avon
Berridge, K. (2001) Reward Learning: Reinforcement, Incentives, and Expectations, The Psychology of Learning and Motivation, 3, Academic Press, New York
Cannon, W. B. (1927) The James-Lange theory of emotion: A critical examination and an alternative theory, American Journal of Psychology, 39, 10-124
Csikszentmihalyi, M. (1990) Flow, the psychology of optimal experience. New York: Harper Collins
Cohen, J. D., Braver, T. S. &.Brown, J. W. (2002) Computational perspectives on dopamine function in prefrontal cortex, Current Opinion in Neurobiology, 12, 223-229
Damasio, A. (1994) Descartes Error: Emotion, Reason, and the Human Brain. Avon: New York
de la Fuente-Fernandez R, Ruth TJ, Sossi V, Schulzer M, Calne DB, Stoessl AJ (2001). Expectation and Dopamine Release: Mechanism of the Placebo Effect in Parkinson's Disease. Science, 293, 1164-1166.
Dickinson, A., Smith. J. & Mirenowicz, J. (2000) Dissociation of Pavlovian and instrumental incentive learning under dopamine antagonists. Behavioral Neuroscience, 40, 468-483.
Donahoe, J.W. & Palmer, D. C. (1993) Learning and Complex Behavior, Needham Heights, Ma: Allyn and Bacon
Fried, I., Wilson, C. L, Morrow, J. W., Cameron, K. A., Behnke, E. D., Ackerson, L. C. and Maidment, N. T. (2001) Increased dopamine release in the human amygdala during performance of cognitive tasks, Nature Neuroscience, 4(2), 201-206
Greicius, M. D. Krasnow, B., Reiss, A. L., & Menon, V. (2003) Functional connectivity in the resting brain: A network analysis of the default mode hypothesis, Proc. Natl. Acad. Sci. 100, 253-258
Holmes, D. S. (1984) Meditation and somatic arousal reduction. A review of the experimental evidence. American Psychologist, 39(1), 1-10
Holmes, D. S. (1988) The influence of meditation versus rest on physiological arousal: a second evaluation. In Michael A. West (Ed.) The Psychology of Meditation, Oxford: Clarendon Press
Homme, L. E. (1965) Perspective in psychology: XXIV Control of coverants, the operants of the mind. The Psychological Record, 15, 501-511
Infante, J.R. (1998) ACTH and beta-endorphin in transcendental meditation. Physiology & Behaviour, 64 (3):311-315
Jacobson, E. (1938) Progressive Relaxation. Chicago: University of Chicago Press
Koepp, M.J., Gunn, R.N., Lawrence, A.D., Cunningham, V.J. Dagher, A. Jones, T., Brooks, D.J. Bench C. J., Grasby, P.M. (1998) Evidence for striatal dopamine release during a video game. Nature, 393. 266-268
Kruschke, J. K. (2001) Toward a unified model of attention in associative learning, Journal of Mathematical Psychology, 45, 812-863
LeDoux, J. (1997) The Emotional Brain: The Mysterious Underpinnings of Emotional Life. New York: Touchstone.
Leuchter, A. F., Cook, I.A, Witte, E. A., Morgan, M., and Abrams, M. (2002): Changes in brain function of depressed subjects during treatment with placebo. American Journal of Psychiatry, 159,122-129
Maclean C.R. (1997) Effects of Transcendental meditation program on adaptive mechanisms: changes in hormone levels and response to stress after 4 months of practice. Psychoneuroendocrinology. 22 (4), 227-295
Martin-Soelch, C., and K. L. Leenders (2001). "Reward mechanisms in the brain and their role in dependence: evidence from neurophysiological and neuroimaging studies." Brain Research Reviews. 36(2-3): 139-49.
McGuigan, F.J. (1976) The function of covert oral behavior in linguistic coding and integral information processing. In Kurt Salzinger (ed.) Psychology in progress: an interim report. Annals of the New York Academy of Sciences, 270, 57-89
McGuigan, F.J. (1978) Cognitive Psychophysiology. Principles of Covert Behavior. Englewood Cliffs, N. J.: Prentice Hall Inc.
McGuigan, F.J. (1997) A Neuromuscular model of mind with clinical and educational applications, Journal of Mind and Behavior, 18(4), pp. 351-370
Montague, P. R., Dayan, P., & Sejnowski, T. J. (1996). A framework for mesencephalic dopamine systems based on predictive Hebbian learning. Journal of Neuroscience, 16(5), 1936-47
Moore, J. (2001) On psychological terms that appeal to the mental. Behavior and Philosophy, 39, 167-186
Panksepp, J. (1998) Affective Neuroscience. New York: Oxford University Press
Panksepp J. (2000) On preventing another century of misunderstanding: toward a psychoetheology of human experience and a psychoneurology of affect. Neuropsychoanalysis, 1(2), 1-21 (on line supplement at neuro.psa.com/pank.htm)
Panksepp, J. (2003) Damasio’s Error? Consciousness and Emotion, 4, 111-134
Raichle, M. E., MacLeod, A. M., Snyder, A. Z., Powers W. J., Gusnard, D. A. & Shulman, G. L. (2001) A default mode of brain function, Proc. Natl. Acad. Sci., 98, 676-682
Schultz, W. (1998) Predictive reward signal of dopamine neurons, Journal of Neurophysiology, 80(1), 1-27
Seligman, M. E. P. (1972) Learned Helplessness, Annual Review of Medicine, 23, 407-412
Selye, H. (1956) The Stress of Life. New York: McGraw-Hill
Staddon, J. E. R. (1993) Behaviorism. London, Duckworth
Schwartz, B. (2004) The Paradox of Choice: Why More is Less. Hopewell, N. J.: Ecco
Taub, E., Ellman, S. J. and Berman, A. J. (1966) Deafferentation in monkeys: effect on conditioned grasp response. Science, 151, 593-594
Teuber, H. L. (1972) Unity and diversity of frontal lobe functions. Acta Neurobiologiae Experimentalis, 32, 615-656
Wilhelm, F. H. & Roth, W. T. (2001) The somatic paradox in DSM-IV anxiety disorders: suggestions for a clinical focus in psychopathology. Biological Psychology, 57(1), 105-140
.
.