The behaviorally-oriented reader might find it useful to imagine pain as a schedule of rewards that becomes available simultaneously with the existing schedules in a given situation; the organism can choose freely among these schedules.   Aversiveness is a schedule that makes available an immediate, brief, intense reward for the first response after a refractory period of fixed duration following the previous occurrence of this reward.   Each occurrence of the reward is followed by a period in which reward is unavailable in most or all modalities.   That is, aversiveness is a fixed interval (FI) schedule of immediate, brief reward in the first link of a chain, followed by a fixed time schedule of time out from reward on all schedules in the second link.  If the capacity for the aversive process to produce spikes of reward regenerates rapidly, that is, if the FI is on the order of fractions of a second, the aversive process as a whole can reward some intrapsychic behaviors like attention but will punish processes  with  longer  latencies  like  motor  behavior.   Furthermore, if responses occur more rapidly than the relevant flicker-fusion frequency, the organism will not be able to identify its separate phases, but only a blend of attraction and aversion.   In our present state of knowledge it would not be profitable to speculate about the events that make this schedule available, or the central nervous system processes that subtend the operant response.

To test this model directly, it would be necessary either to dissect out the distinct rewarding and punishing components in known aversive stimuli or to create aversive stimuli from known rewarding components.   The former is probably beyond the present ability of physiological psychology.   The latter seems to have been done, although on a slower time scale than the one I have proposed for pain: When pigeons are given an increasingly lean fixed-ratio schedule of food reward, they come to actively prefer time out from the schedule to the schedule itself (Appel, 1963; Azrin, 1961; Zimmerman & Ferster, 1964).  That is, the mere chance to perform an operant behavior has become aversive, even though subjects perform this behavior when it is possible, or perhaps because subjects perform this behavior when it is possible.   In the same way, it can be argued, organisms are briefly motivated to perform the operant that generates motivational-affective pain, but most of the time prefer a schedule which does not offer such an opportunity.

This argument does not lead to the conclusion that pain is 'really pleasure,' nor will it shed light on the puzzling finding of motor behaviors that seem to be maintained by punishment (e.g.  Dweyer & Renner, 1971).   It does not require that the rewarding component of aversive events be conscious nor the attention to the events deliberate.   However, some awareness of an alluring aspect to aversive events does seem to be discernible in our culture.

The idea of fatal temptation to pay attention is common both in mythology and everyday experience: Orpheus lost Eurydice because he could not resist looking at her; Pandora unleashed all the world's troubles because of curiosity (Bulfinch, 1948, pp.  201-204, 12-15); and Lot's wife was turned into a pillar of salt because she could not help looking at the destruction of Sodom (Genesis 19: 12-29).  Similarly, our cliffside walker has a hard time not looking down, and drivers who cannot avoid noticing the proximity of death at highway speeds may become unable to drive.  E.B.  Titchener, listing those stimuli which people "cannot help" paying attention to, included "newspaper accounts of fires and murders [which] have a morbid fascination for us (1910)." Unpleasant rumination about traumatic events of the past is widespread, occurring in 71% of elderly subjects in one study (Tait & Silver, 1989).

The hypothesis that unpleasant stimuli lure us rather than push us should not be hard to accept.   Similarly, there are many familiar behaviors which we find unpleasant and can withhold, but only with the greatest effort: biting a canker sore, rehearsing a bygone humiliation, or reacting with arousal when we hear a dripping faucet while trying to go to sleep.  Indeed, the combination  of  attention-drawing  and  behavior-deterrence characteristic of physical pain are shared by a number of other aversive processes.   These hinge on behaviors as brief as the person's very notice.   Phobias are the most important category, the  participatory  nature  of which is  shown  by  their responsiveness to behavior therapies that give patients practice in resisting the urge to panic (reviewed in Clum, 1989).   Tinnitus (ringing in the ears) seems to be another example of a perception that can be reduced by training in structuring attention (Ince et. al., 1983).  There  are many other 'feelings' which are experienced as 'happening to' the person but which can be cultivated or, conversely, starved out by some kind of practice.  If the person experiences them as happening without his participation and regrets them from their first appearance, they should be regarded as temporary preferences in the pain range of duration.

 

Repression can be seen in terms of simple information processing.   When deciding whether to pursue a given activity, a person does not call up all his knowledge of it at once, but begins with a label by which he has categorized this knowledge (Shiffrin & Schneider, 1977).  He is apt to evaluate his options for further information processing according to the likely payoffs for these options, perhaps arrayed in what has been called a 'sentry matrix' (Bruner et. al., 1956, p. 75).  If he has categorized his knowledge of the activity according to its impulsiveness, and the capsule he recalls first tells him that the activity is indeed impulsive, he will know that further review risks revealing it to be imminently available; he may thus be motivated to stop the review at that point.

   Such avoidance of further information represents the act of repression.   Of course, if he estimates the risk of discovering its availability to be high already, that very estimate may have changed his preference in favor of further review, and the repression will have failed.  Furthermore, if the capsule contains information that further review is apt to be unpleasant (using what Williams et.al. call 'affective salience'-- 1988, p. 171), he may decide in his short range interest not to pursue it further.   He will then be using repression for its other defensive purpose, maintaining short range comfort.   Whatever its purpose, if the person were able to report his decision to avoid further information it would be called 'suppression' rather than repression, a distinction that is not important for this discussion.

   Freud believed that all repression was pathological, and particularly apt to lead to 'hysterical' symptoms (Breuer & Freud, 1895).  Normal subjects who say that attention-diverting tactics make sense as impulse controls are indeed more likely to be female, but no more likely than other subjects to report 'hysterical' personality traits (Ainslie, 1987a).

   Before Freud's time forgetting about one's impulses was not only a spontaneous  behavior but the  recommendation  of professional advice-givers.   For instance, the writer of a book called Right and Wrong Thinking and Their Results advised the reader to "avoid discordant thoughts," by distraction if possible and if necessary by "the rule at Donnybrook Fair:  'whenever you see a head, hit it.' The least is not too small to be terminated if it is wrong (Crane, 1905, p.115)." This would seem to be an attempt to describe sheer suppression.   Behavioral writers even today advocate 'stimulus control' as a useful way of avoiding impulses (Kanfer, 1975, pp. 309-355; Goldiamond 1965).  It is common lore that "if you speak of the Devil, he'll appear."   Patients have described to the author being able to 'fight off' panic attacks, dissociative episodes, and even epileptic seizures by vigorously directing their minds away from the feeling that these things were about to occur.

   Denial, the misinterpretation of what one notices, may also serve this purpose by distorting information about environmental opportunities for impulses.  It is often advocated in the form of hypnosis, particularly against preferences of brief duration like pain  (Hilgard & Hilgard,  1975) and panic,  although its effectiveness against panic has been limited (Schneck, 1954; Wolberg,  1948).   However, psychoanalytic writers recognize denial mostly as serving to avoid painful perceptions rather than as a precommitting device.

 

Early inhibition of emotions is probably a powerful means of precommitment, although this device costs whatever reward is dependent on that emotion for its consumption.  For instance, the person who controlled his sexual temptations by the early avoidance of sexual affects might run the risk of losing his capacity for sexual enjoyment.

   A person can also decrease the attractiveness of  a particular activity by cultivating a contradictory emotion.   For instance, when entering a situation which he expects to provoke unwanted tender feelings he might forestall these feelings by summoning his rage at the earliest opportunity.   Conversely, if he is worried about rage he might cultivate tender feelings.  Examples of this device have been discussed under the name of reversal of affect (A.  Freud, 1966, pp. 29-40; Freud, 1914, pp.  126-127).   This device has also been proposed by behavior therapists, as we shall see presently.

   Reversal seems to represent a special case of general strategy: finding activities that reduce one's appetite for, or increase one's appetite for the alternative to, a particular reward.  This general strategy has been called reaction formation (A.  Freud, 1966, pp. 37-38; Freud, 1926, pp. 157-158).

Redefining personal rules: the role of bright lines.  These three requirements-- adequate size, adequate clarity,and criteria which do not depend on the person's own behavior-- are all that should be necessary for the success of personal rules made in advance and maintained without modification.   However, in a changing environment, a person is apt to find that he has committed himself to forego an unexpectedly large amount of reward.  Such a commitment will lead him either to break his rule or to forego more reward than he had expected to in order to follow it.   He might be able to get more reward out of the situation if he could redefine his rule without undermining its effectiveness.

For instance, our dieter might find that he had committed himself to forego a meal with great  sentimental value (Thanksgiving dinner, say), or to offend a host by refusing food, or simply to avoid a kind of food which he liked more than he had thought.   There are any number of ways he could make the necessary loopholes in his diet: calling it off on major holidays, or in company, or when the food would be wasted if he did not eat it, or just this once, and so on.   It is almost always possible to formulate a principle that will grant an exception in the case at hand; but that very fact makes indiscriminate redefinitions fatal to a personal rule.

If the dieter is going to redefine his rule in such a way that he keeps his expectation of losing weight, the occasion he uses must not only be rare but must also stand out in some way from other possible occasions which are not rare enough.  He must detect some feature that separates his current, adequately rare excuse from other excuses that would be available too often.   If for him Thanksgiving stands out against Sunday or a cousin's birthday, if he can name only a few specific hosts who might be hurt if he refused large portions, or if an infrequently encountered dessert actually has a unique place in his heart, only  then might he make an exception without losing the credibility of his rule.   The interest based on the smaller, earlier reward finds countless ways to say, "just this once;" facile reclassification of behaviors that were originally forbidden is called rationalization (Hollitscher, 1939).   The long-term interest must defend itself by arguing, "This, too, is a matter of principle." Ultimately the decision between the two depends on the person's guess about whether a particular exception will in fact lower his expectation of generally adhering to his rule.  He may even accept some probable lowering of this expectation, so long as he is left with enough to deter most lapses-- Calculation of that quantity is an internal equivalent to estimating how many defections there can be in a multi-person prisoner's dilemma without making cooperation generally unprofitable (Schelling, 1978, p. 217ff.).

The problem of defining exceptions is inherent in all real life examples of repeated prisoner's dilemmas.   The cooperation between the countries in the choice about poison gas  is threatened by the possibility of choices that are marginal in their aptness to be seen as precedents, such as the use of gas against a third nation or the use of an explosive that happens to give off toxic fumes.   A country might hope to engage in such marginal behaviors without being perceived as having betrayed its tacit agreement to cooperate.   Because of this hope, there is substantially more risk that the country will engage in them and find it has hoped falsely than that it will commit an unambiguous betrayal.

Thus the tacit cooperation that limits warfare depends on the clear identifiability of defections.   This in turn depends not only on the astuteness of the rules for cooperation, but also on what could be called the topography of the choice situation-- the physical features of the various alternatives that might be chosen.

The availability of boundaries which cannot be moved just a little bit is crucially important to the long-term interest.   Activities like smoking and drinking have such a line in an obvious place, that is, between any indulgence and no indulgence; but people who eat too much or spend too much money cannot completely give up these activities, and so must find some way to make a single diet, or budget, stand out from all the others to which they are apt to retreat under pressure.   Lawyers call such a unique boundary a 'bright line.'  The concept is familiar to people whose profession it is to negotiate between interests in the larger world.  It expresses why countries blessed with unique boundaries like a mountain range or a river without large tributaries have fewer wars than countries just set out on a plain.

Whether people, or nations, in a limited war situation must depend on a bright line to maintain cooperation, or can use less prominent lines to gain more flexibility, depends on factors like their history and skill in that situation and the amounts at stake.   For instance, war between great powers may have been prevented since World War II by the widespread belief that even skilled policy-making could not restrict it to conventional weapons.  Thus their very history of failing to avert the escalation of wars, added to the new threat of nuclear destruction, may have deterred them from venturing beyond the bright line between some war and no war at all.   Similarly, alcoholics  find  that they cannot engage  in  'controlled drinking'-- following somewhat arbitrary rules to stop after two drinks, or three, or when they feel high, or when their spouses say they have had enough-- and are advised by  Alcoholics Anonymous to regard themselves as 'helpless against alcohol.'

To be helpless means one cannot use his will power flexibly in this area, that is, cannot successfully choose one principle of drinking or another, but can only hope never to be lured across the bright line between some drinking and no drinking.   Strictly speaking these alcoholics are still using will power, in that their choice is constrained by the logic of a repeated prisoner's dilemma among successive motivational states; but since this constraint is determined by a single bright line in their environments rather than their own activity in defining what their rules will be, they may indeed feel helpless (v.i.  Ch7).  People for whom drinking is not so rewarding, and whose wills have not lost their credibility in the area of drinking, are able to obey less prominent lines or even their spontaneous preferences without losing control.

Robert Frank has recently pointed out that a morality higher than the simple prudence that requires one not to risk getting caught may still be consistent with strict self-interest (1988, pp.  71-95).  The logic is exactly that of Alcoholics Anonymous:  Because of the tendency to overvalue present gratification of one's own particular vices, a person who tries to indulge them in a controlled way according to rules for prudence will often find loopholes that lead him beyond prudence into behavior for which he will be caught and blamed.   The person who draws a line farther from his impulses, for instance to be virtuous for the sake of virtue, has the same improved chance of avoiding his impulses as the alcoholic who perceives himself to be helpless against alcohol.  This may be the only person who succeeds enough at being prudent to keep a reputation for virtue.   Unless he sincerely renounces the relevant vice, the person may be like the 'dry drunk' who is just waiting for a loophole.

It is worth noting a parallel that has been proposed with the evolutionary competition of organisms. Some authors draw an analogy between the selection of organisms and the selec¬tion of behaviors within an organism (Gilbert, 1972; Skinner, 1981). Vaughan and Herrnstein have suggested that the two processes are formally identical (1985). There are indeed many similarities between them, but also two major differences: Organisms (or their genotypes) are literally separate enti¬ties, and their selection is based on the complex hap¬penstances of their environment, most of which are not in turn subject to natural selection. (For instance, the cold wave that selects for cold-adapted organisms does not do so because that selective function improves the cold wave's own chance of survival.) An organism's behaviors, on the other hand, are related by a common dependence on the activity of that organism's reward center (or interacting set of reward centers), which at least partially unites behaviors that are separated in time; that selective function has in turn been shaped by the natural selection of organisms to serve a coherent purpose, the survival of that organism and its progeny. (For instance, seeking warmth in the cold is selected for by the organism's reward mechanism because that reward process has been adaptive.) The evolution-shaped function by which the reward process discounts delayed events is apt to be different from the unselected mathematical function that simply describes the survival of the fittest over time.

The degree to which a rule for behavior takes on a compulsive, externally imposed, quality may partly depend on how actively the person set up the private side bet that enforces it ++ When I wrote this I was still referring to personal rules by their mechanism of recruiting additional motivation—private side betting ++.   Private side bets may form spontaneously whenever a person notices that his individual choices to resist or yield to minor temptations are members of a larger class.   He begins to refer his recurring choices to crude, possibly unnamed categories having to do with when he will give in to the urges for procrastination, lying, or wasting money, as well as for such unwelcome but controllable emotions as panic, disgust, and anger.  For instance, he may notice that if he does not get up as soon as his alarm rings, he loses his ability to get up on time.  After some experience with this problem, he may find that he 'cannot' stay in bed once the alarm has rung.

   Likewise, for a person who is susceptible to panic, fear is never far away.   Where he can resist the urge he must resist it, lest he take warning from the failure and abandon another area of functioning to his fear.   He is unlikely to describe this requirement as a rule, but just as using some kind of effort to fight the fear; those cues which, according to the above analysis, lead him to abandon the effort by predicting its failure, he will simply describe as objects of fear.  Conversely, if some coincidental circumstance has been accompanied by a period of confidence, the person may 'superstitiously' begin seeking more of this circumstance in order to ward off panic.   But if the superstition were merely a guess about his likelihood of panic, it would soon be disproven and abandoned.   A durable superstition functions as a rule, eg.  "Don't panic when your spouse is beside you" or "when your horoscope is good" or "when you have performed the correct ritual." The stake of such a rule is the person's expected freedom from panic in the area that is in doubt.   He has gained power over an urge, but at the expense of being bound to his particular talisman-- Take it away, and what might once have been an open choice becomes helpless abandonment to the relevant vice district.  ++ Same concept as lapse district {Rules magnify lapses} ++.   In either case, the effect appears to be magical.

   Many phenomena called supernatural are probably examples of an unrecognized private side-betting process, as I have argued elsewhere in the case of oaths (Ainslie, 1975, p. 483): An oath defines a category of behaviors-- "things sworn to"-- that the person thereafter performs or fails to perform as a set, with the good will or wrath of the sacred entity sworn upon explaining the consequent double-or-nothing quality of the person's motivation.  As we have just seen,  Max Weber ascribed the behavioral consequences of Calvinism to a similar mechanism.   Without a concept of private side betting, people naturally postulate a supernatural version of ordinary public side bets, a perception that in no way diminishes the effectiveness of the bets.

   A personal rule that has gravitated into being as in the counterphobic example may seem to be more alien than one which the person set up by a deliberate oath or resolution.   However, most rules in most people arise through just such mindless practice without seeming like alien forces. The compulsive quality of a rule may depend more on the extent to which the person is actually bound by it.   If the impulses that necessitated the rule seem disastrous, if the person knows no other defenses that would be adequate to forestall these impulses, and if he does not see safe ways to redefine the rule to make it adapt to his changing needs, then he will be apt to say that he is compelled by a force outside his own will.

   For instance, if he saw stepping on a crack as symbolic of an urge to break his mother's back, and feared that his anger at his mother, once aroused, would be so compelling that any expression of it must be forbidden, he might become 'unable' to step on cracks.  He would be said to have developed a compulsion.   The actual force of the compulsion would be the aggregate effect of the whole category of long-term rewards at stake, that is, everything he feared losing if he hurt his mother.

 

  Many activities require a preparatory process in order to be fully rewarding, sometimes in order to be rewarding at all.   A person must get into the right mood to enjoy an evening of farce, or an argument, or a square dance.   Even for concrete rewards-- especially for concrete rewards-- a process of 'appetizing' is important.   'Appetite' is sometimes used as a synonym for taste, but 'taste' does not also denote this transient state of readiness to be rewarded, as in "his appetite was aroused." Appetite in this second sense is a behavior in its own right, which is performed where it has been rewarded and which may or may not be followed by consummatory activity.  It is related to taste in a way that may account for changes of taste in the service of self-control:

Where the rewarding power of an activity depends on a preparatory appetite, the need to arouse this appetite imposes a delay between choice and reward.  Such a delay is apt to keep the activity from being overwhelmingly tempting, as long as the appetite has not been aroused.   In that case, the person can supplement his personal rules with a form of emotion control: early avoidance of the appetite.   At a moment when a person can initiate an appetite for a short range reward, that reward may be distant enough that his knowledge of its consequences would deter the appetite.   The person will just not 'work it up,' and thus may maintain his long range preference.  If he regu­larly inhibits appetites for a particular activity, he will be said not to have a taste for it.   He can thus restrict his tastes to only some of the activities that could be rewarded by the underlying drive.

 

[Complete obedience to a rule against arousing a particular appetite can eliminate the taste for it. -modified May 21, 2019]  If this rule is strong enough to convince him that there is no likelihood of his indulging in the activity, this fact will motivate him not to keep considering it as an option.  He will stop imagining its rewards, and will avoid initiating a futile appetite for it, unless something in the appetitive process itself is sufficiently rewarding to repay the effort.  A re­formed smoker who is sure that he will not smoke again will usually not think about cigarettes, and eventually reports that he has lost his craving for them or even become disgusted by them; but a person may find it rewarding to fantasy a sexual adventure that he is equally sure he will not undertake.

Where an appetite is rewarding in its own right the person may need an additional rule prohibiting not only the con­sumption of  a particular class of rewards, but also the preparatory process that makes them rewarding, lest enter­taining the appetite would bring reward for the forbidden activity too close to be resisted.   That is, simple foresight may not suffice to inhibit appetites that are tempting in themselves; personal rules may be necessary.  Insofar as these rules succeed his taste will have changed by the first mechanism, avoidance of appetite.  Of course, if his rule against the consummatory activity is weak enough that it some­times occurs, the appetite for it will be rewarded and is also apt to occur even if this appetite is not rewarding in its own right.

Thus the application of self-control tactics to either consummatory activities or appetites for them may affect the experience of taste.  It is not hard to find examples of tastes that have been edited down in the service of self-control from what their underlying drives originally made possible.   For instance, all societies regard some sources of reward as taboo or unclean, and their members have effectively lost their capacity to be rewarded by these sources (Douglas, 1966).  Contact with a tabooed object has even been used as a punish­ment.   People in modern society usually rule out dogs or other pets as food, and members of the nuclear family as sexual objects, although both are biologically able to deliver the relevant rewards.  The dangers attached to them are clearly of the moral sort.   However, the rules against using them in these ways are experienced as lack of desire for them or even as disgust at the idea of using them in this way, rather than as restraints against a desire; the tastes themselves have changed.

As Lazarus has argued in a case of negative tastes, the phobias, the attempt to trace all tastes to simple pairings of their objects with reward or punishment strains credibility (1972;  v.s. p.##15; evidence confirming this view in Murray & Foote, 1979).   He distinguished those phobias formed by contiguous punishment and those formed by ideational processes, a distinction reaffirmed by Wolpe (1981).   An example of the latter could be the a priori deduction that an object is un­clean or taboo, in which case classical conditioning would clearly founder as an explanation.   Although tastes are experienced as beyond the control of the will, they nevertheless must be regarded as motivated processes that can sometimes shift because they exemplify a larger principle of impulse control.

It has long been a psychoanalytic tenet that people regulate their appetites and thus form tastes with a view to augmenting their impulse control (Salzman, 1965).  The concept of personal rules governing the generation of appetites seems to coincide with the psychoanalytic concept of 'bound cathexis.'  Such cathexis, perhaps better translated as 'importance,' is given to an object by a person, but thereafter inheres in the object in an inertial way so that it cannot be withdrawn easily (Freud, 1916-17, pp. 348, 422).  For example, a man who per­ceives sexual activity with women to be a trap may forbid him­self a sexual appetite in the presence of women.   However, if he notices a line which divides some less dangerous sexual activity from the sexual activity he fears, he will be motivat­ed by the whole force of his sexual drive to seek that crite­rion (Fenichel, 1945, p.  330; Freud, 1938, pp. 202-203).   Such a criterion might be the presence of women's under­wear, or pre-pubescent girls, or boys, or some situation in which women seem less threatening, such as exhibitionism.   The criterion he adopts for permitting a sexual appetite will be the hinge of a personal rule, the stake for which will be all that he expected to lose if he succumbed to his original sexual impulse.   This stake could be described as the energy with which the criterion is cathected (Freud, 1923, pp. 42-47).  The common perception of this situation will be that he has a taste for the sexual stimuli on the permissible side of the criterion.

It is not obvious why personal rules sometimes permit the generation of appetites for temporarily preferred  rewards, producing the experience of a desire opposed by either a deliberate rule or a compulsion, and sometimes forbid the appetite itself, producing the experience of indifference or revulsion.  Why, for instance, do some people remain attracted to a sexual experience they have forbidden to themselves, while others cease generating this sexual appetite or generate dis­gust?  Perhaps forbidding an appetite itself produces greater safety against the impulse, but at the expense of some harmless reward that cannot be consumed in the absence of the appetite.   Ruling out an appetite would thus be a stronger but potential­ly more costly defense than simply ruling out its correspond­ing consummatory behavior.