The evolutionary stability of a bi-stable system of emotions and motivations in species with an open-ended capacity for learning

From Point Omega Research
Jump to: navigation, search

This article is a copy of the original article by Popko van der Molen. That paper was written as an answer to questions put forward by Ian Mathie, University College, Cardiff, and Bill Livant, University of Regina, Canada, at the International Symposium on Reversal Theory, University of Wales, Sept. 2 - 4, 1983. Their stimulating remarks during that conference are gratefully acknowledged.

Modified versions of this paper have been published in Acta Biotheoretica (Molen, P.P. van der (1984): "Bi-stability of emotions and motivations: An evolutionary consequence of the open-ended capacity for learning") and in J.Wind and V.Reynolds (Eds., 1986): "Essays in Human Sociobiology". The implications of this theory for psychotherapy and growth-psychology have been published in Apter, Fontana & Murgatroyd 1985 and in the British Journal of Guidance and Counselling (van der Molen,1986).

(**) Abstract

- One of the highest evolutionary achievements is the open-ended capacity for learning. This is the ability to acquire a behavioural repertoire which is specifically tailored to the environmental situation(s) an individual happens to live in. This capacity is best exploited if the individual's behavioural organization causes any surplus of energy to be invested in expanding and refining the repertoire, and adapting it to prevailing circumstances. Adaptations of the repertoire are most likely to have survival value if applicable to emergencies and other situations causing high arousal. Experience and skills are therefore maximizing fitness if acquired, and subsequently used, in arousal-evoking, and often risk involving, situations.

Entering high-arousal situation, either voluntarily or involuntarily, however, may be harmful to the individual if the resulting state of high stress lasts too long to allow proper (neuro-)physiological functioning, or if too much risk is involved. An open-ended learning capacity is therefore maximally adding to survival if paired to two distinct tendencies:

  1. a tendency to seek high-arousal evoking situations whenever surplus energy is available; and
  2. a tendency to seek arousal reducing situations as soon as an emergency occurs or as soon as the surplus energy is exhausted.

This suggests that a bi-stable "telic/paratelic" system of preferred levels of arousal, as described in Apter & Smith's theory of motivational reversals (Apter 1982), can be considered an Evolutionary Stable Strategy (E.S.S.), as compared to homeostatic systems of arousal and motivation.


(***) One of the most recent evolutionary achievements is an open-ended capacity for learning. This is the ability to acquire a behavioural repertoire which is specifically tailored to the environmental situation(s) an individual happens to live in. It will be argued here that this ability is highly enhanced by a bi-stable organization of motivation as described by Apter & Smith in their "reversal theory"(see Smith & Apter 1975; Apter 1976, 1982; Apter & Smith 1976abc, 1977, 1979; and Apter, Fontana & Murgatroyd 1985).

In fact, it will be shown that the predisposition for such a bi-modal antagonist system of emotional and motivational reversals constitutes the basis of behavioural flexibility Pelt (this volume) is referring to. The present paper therefore deals with the dynamic structure of the biological substrate of what may be called "memes" (Dawkins 1976) or "culturgens" (Lumsden & Wilson 1981). It deals, so to speak, with the interface between "genes" and "memes", between the "hardware" and the "software", as has occurred in man. Here we will deal with the motivational mechanisms taking care of the acquirement and the selection of "software" at the individual level.

(For returning to the "Guided Tour", click here. For more detail, continue.)

Reversal Theory

(***) Let us first have a look at what this "reversal theory" is all about. Apter (1982) introduced his reasons for proposing a model of motivational bi-stability as follows and as schematically pictured in Fig. 1.

Let us start by reflecting on a number of different situations in which different levels of arousal, high and low, can be experienced, and see if any pattern or structure can be discerned.
First of all reflect, if you will, on what it is like to be in a dentist's waiting room, about to have a filling. Focus on particular on the kind of arousal which you feel. Now imagine that you are soaking in a hot bath after a hard day's work, and again try to conjure up what the arousal which you experience in this situation feels like. Next, make believe that you are waiting for a bus which is taking a long time to come; you are not in a hurry so that there is no pressure of time, but you have nothing to read or take your attention. Finally, imagine yourself to be in a cinema watching a thriller film and the film has a particularly tense point. I thing you will agree that the arousal feels different in some important sense in each of these cases.
For one thing, of course, the amount of arousal which you feel will be different. If you are like most people, the level of arousal which you experience is likely to be high to some degree or another in two of the cases (waiting for the dentist and watching the film), whereas in the other two cases (waiting for bus and in the bath) it is likely to be rather low.
A second obvious way in which the arousal-experience will be found to differ is in terms of its pleasantness or unpleasantness. Again, two of the situations listed are likely to be found to involve unpleasant arousal (waiting for the bus and waiting for the dentist). Pleasantness and unpleasantness are not, therefore, simple reflections of the amount of arousal.
In terms of just these four examples, high arousal can be pleasant or unpleasant, and so can low arousal. In fact, there would appear to be at least four types of arousal experience. This is corroborated by the fact that there are four widely-used arousal words in everyday language which relate exactly to these four types: anxiety (unpleasant high arousal), excitement (pleasant high arousal), boredom (unpleasant low arousal) and relaxation (pleasant low arousal).
Figure 1. The very existence of colloquial labels for our emotions like relaxation, boredom, excitement and anxiety illustrates emotional/motivational bi-modality.

Apter & Smith (1979) have generalized these and similar findings into a simple but comprehensive theory. One of the fundamental postulates of their theory is

[...] that certain psychological processes, especially certain motivational and emotional processes, exhibit bistability rather than homeostasis (i.e. unistability). Switching from one stable state to the other in a bistable system can be referred to as a "reversal" (hence the name of the theory) and may be brought about by a number of different factors. (See fig.2)
Figure 2. At certain times the individual seeks high arousal which is then felt as pleasant when achieved ("excitement"); at other times he reverses to a state in which he seeks low arousal, at which time high arousal is felt as unpleasant ("anxiety"). In the former case low arousal is felt as unpleasant ("boredom") and in the latter case low arousal is felt as pleasant ("relaxation"). (After Apter & Smith, 1979).
Figure 3. Each hypothetical curve represents the relationship between arousal and affective tone for one of the two stable states. These two states are labels "telic" (goal-directed; from the Greek word "telos" = "goal, end, close") and "paratelic" (behaviour-directed) respectively. (After Apter & Smith 1979, with permission).

Apter & Smith (1979) explain that the point of presenting the above figure 3 is to illustrate that the relationship between arousal and anxiety may not be linear but bi-variate: high levels of arousal may provoke anxiety in the telic state but may be regarded as both exciting and pleasurable in the paratelic. Thus this figure implies that, when the homeostatic nature of certain psychological processes is rejected in favor of bi-stable models of such processes, human action may be seen to be far more complex. Evidence of these relationships can be found in Apter (1976, 1982) and in Apter et al. (1985).

It should be emphasized that the telic and paratelic states are self-perception-determined. That is, a person can be said to be in one or other of these states only as a result of the way in which he sees his own actions; it is not the actions themselves which identify the operative state an individual is experiencing.

To illustrate this point, consider an individual driving a car at 130 mph. This action may be regarded by the individual performing it as telic if he is driving at this speed in order not to miss an important appointment very relevant to an essential business transaction. Alternatively, the individual may perceive his action as paratelic if he performs it because driving at high speed is thrilling and exciting. This point---that these states are self-perception-determined and may be examined only by reference to the individual's perception of his own action---is crucial to an understanding of the theory of reversals.

This and other features of the theory involve a development and systematization of the notion (Schachter & Singer 1962) that particular emotions derive from the conjunction of a particular arousal level with a particular cognitive interpretation.

Difference with homeostatic theories

(***) The view of Apter & Smith on the relationship between felt arousal and affective tone contrasts sharply with Freudian theory, with 'drive-reduction' theory, and with optimal arousal theory, which are all homeostatic theories. The basis of Freud's earlier theory of motivation is the constancy or 'stability' principle, which he took over from Fechner (1873) and restated in the following terms: "The mental apparatus endeavours to keep the quantity of excitation present in it as low as possible or at least to keep it constant" (Freud 1920). This principle dates back to the beginning of Freud's psychological work, and as early as 1888 he was writing of a "stable amount of excitation" (Freud 1888). Apter (1982, p. 129) suggests that it forms the basis for Freud's pleasure principle, the idea being that 'unpleasure' is avoided if excitation (or 'tension') is kept constant or reduced to a minimum.

In a similar way, reversal theory differs from, and is an essential addition to, other homeostatic theories of motivation; e.g., Lorenz'(1950) hydrolic models of (aggressible) behavior; Hull's (1943) learning theory which assumes that all reinforcement, and therefore all learning, depends ultimately on the reduction of primary homeostatic drives; Zuckerman's theory (1974, pp. 82, 136) on "sensation-seeking"; and Festinger's (1957) theory of "cognitive dissonance".

From the perspective of reversal theory, then, most theories of motivation up to the present time have been at best little more than half theories: they have in the main dealt in their different ways with motivation in relation to the telic system, but not to the paratelic. (Apter 1982, p. 133).

The alternating antagonistic telic and paratelic tendencies may be expressed at various levels and areas of functioning (see Table 1).

Table 1. The Telic and Paratelic mode in relation to Means-Ends, Time and Intensity

On explorative behavior and the familiarity/novelty opposition Smith & Apter 1975, pp. 10,11 write:

In the telic system, familiarity is the dominant member of the pair and novelty is the dominant member in the paratelic system. That is to say, in the telic system the organism searches for security and safety in order to reduce arousal; in the paratelic system it searches for novelty and surprise in order to increase arousal. However, calling exploration a drive as is now done widely, following the work of Butler 1953, Berlyns 1960, and others, implies that to discover new things is arousal reducing. Our claim is the contrary: namely that such discovery increases arousal and, in the paratelic state, this is in fact pleasurable. This idea therefore constitutes an attack on the assumption that all behaviour is governed by principles of drive-reduction.

Whereas this "reversal theory" cannot be denied intrinsic elegance, it is not exceptionally attractive in terms of parsimony and simplicity, compared to models of drive-reduction, of homeostatis, and of optimum-arousal theory. As may be clear from the foregoing, Apter & Smith defended the introduction of their theory quite straightforwardly on the grounds of fitting better to the data.

Below I will present another argument in favour of adopting reversal theory, an argument in terms of evolutionary stability of behavioural strategies. The evolutionary stability of the reversal system is linked directly to the existence of an open-ended learning capacity. But in order to see how a "reversal system of antagonistic motivational modes of behaviour" ties in with open-ended learning, and how it can be stable evolutionarily, we will first have to take a closer look at the organizational requirements and implications of the capacity for open-ended learning itself.

Requirements of open-ended learning

(***) If an open-ended learning capacity is present, it is of course best exploited if also a behavioral program is included which causes any surplus energy to be invested in expanding and refining the behavioural repertoire and in adaptations to prevailing circumstances. Surplus energy is to be understood here as the basis of physical and psychological preparedness for activity, i.e., after the various immediate goals, related to physiological and social needs have been fulfilled. There are, of course, differences between the various levels of energy, e.g., between the physical and the psychological level, but it is not necessary to differentiate here. Surplus energy on any one level may facilitate paratelic activity on the level in question. Even when being dead-tired physically, one may very well derive pleasure from arousal on a purely mental level, and the other way around.

Furthermore, it matters what sort of experience and skills are gathered, and in what sort of situation. Most crucial to the individual are those experiences and skills that are applicable to emergencies and to other situations of vital importance. Skills are especially needed when much is at stake and when high risks are involved. Situations of vital importance in which much is at stake, are likely to be highly arousal inducing; and conversely, experiences and skills, learned in highly arousal inducing situations, are more likely to be applicable or transferable to emergencies and other vital situations than are experiences gathered in low-arousal situations. Whenever surplus-energy is available to be invested in the learning process, it seems therefore most advantageous if this energy is invested in high arousal inducing situations.

On the other hand, high arousal inducing situations often imply relatively high risks and a relatively high probability of emergencies turning up. If one seeks excitement, one might get more arousal than one would appreciate; and this is felt then as anxiety. What, after all, is exciting (arousal inducing)? Exploration, the unfamiliar, experimentation with situations and social roles, play, spontaneity, etc. are; and such behavioral tendencies are likely to lead sooner or later to unforeseen problems or to outright emergencies. Emergencies and other problem situations evoke high arousal and other behavioral adaptations useful for immediate and short-lasting top-performance. But such adaptations for emergency peak performance are harmful to the individual if they last too long. A neuro-physiologically strong activation of the sympathetic system has to give way after some time to the activation of the parasympathetic system; if not, lasting damage to the individual's organs will occur and physiological balance will not be restored.

It is therefore essential for an individual to stop seeking arousal (excitement) as soon as exploration, play, spontaneous experimentation, or any other excitement-seeking behavior has run out of hand and an emergency had arisen (anxiety), or as soon as his surplus-energy has been exhausted (fatigue).

In summary, the survival value of an open-ended learning capacity is maximized, if paired to two tendencies:

  1. The tendency to seek high-arousal-evoking situation whenever surplus-energy is available; and
  2. the antagonistic tendency to seek arousal-reducing situations as soon as an emergency occurs, or as soon as the surplus-energy has been exhausted.

Such elaborations may seem a bit superfluous and self-evident, but combination of these two requirements in fact implies a departure from optimum-arousal theory, and from principles of drive-reduction.

In fact, these requirements are met perfectly by a bi-modal organization of motivation as described by Apter and Smith's hypothesis. It is therefore concluded that, given the possibility of a telic/paratelic system of antagonistic motivational states, an open-ended learning capacity without such a system can never be an Evolutionary Stable Strategy (E.S.S.) (Molen 1984)1985.

Function of the motivational states in the process of learning

(***) Having concluded that a reversal system of telic and paratelic antagonistic motivational states is a behavioural adaptation to maximize the effect of an open-ended learning capacity, we still need to find out how the learning process works as seen in the light of that reversal system.

One commonly occurring sequence of changes between different types of experienced arousal is depicted in Figure 4b (see Apter, 1982, p. 100). It is important to note that a reversal from relaxation 1 to boredom 2 (as indicated in Fig. 4b) is not just dependent on a quick and sufficient decrease of the arousal level, but that some time is required during which the state of relaxation must last and grow deep enough for the telic state to become satiated. In other words, after anxiety has been overcome successfully by effective attempts to reduce the level of arousal, the individual needs some time to restore mental and physical equilibrium to the extent as is necessary for the paratelic state to emerge again. In fact, as was explained in Section 4, the shift to the paratelic state can be assumed to depend on the availability of surplus-energy which can be invested in gathering experience and information.

Furthermore, reversal from excitement 3 to anxiety 4 (Figure 4b) is not always simply facilitated by a gradual increase of arousal during the paratelic, arousal-seeking mode of behavior, but occurs often very suddenly because of an emergency situation (reversal by contingency). After all, paratelic behavior is non-telic, i.e., goal-less and open-ended. As was pointed out above, an individual in a paratelic state is therefore likely to encounter novel, unfamiliar situations from which emergencies may arise all of a sudden.

The paratelic behavior state is, by its open-ended and exploratory nature, important for the process of expanding the behavioral repertoire. However, experiences gathered in the telic state are also of vital importance. In that state the effectiveness of acquired skills and experiences is further tested and modified during serious attempts to escape from anxiety-evoking emergencies.

It is therefore not just the paratelic, behavioral expansive, exploratory state which is important for the process of learning, but rather the sequence of changes as depicted in Figure 4, including all the different aspects of striving and playing, agony and pleasure, involved. From this it follows that if this sequence of changes in moods and strategies is hampered, for instance by a persistent inability to attain relaxation when in the telic state, the process of learning is essentially impeded. And such an inability to attain relaxation may stem from internal factors (unskills) as well as from unfavourable environmental factors.

Figure 4. Reversal system of antagonistic motivations. a. "Telic" and "Paratelic" moods may be considered as antagonistic motivational states. b. The dynamic process of motivational reversals steers the acquirement and the processing of experiences. A common sequence in this dynamic process is depicted here. If an individual can muster sufficient skills to reach and consolidate relaxation after having arrived in a telic state, his state of relaxation will after some time reverse into boredom, which implies reversal from a telic to a paratelic frame of mind. After having succeeded in the paratelic state to become sufficiently strong stimulated and aroused as to fulfil the—paratelic—desire for excitement, the strongly stimulating situation in question may eventually prove to cause too much arousal to bear. At that moment excitement reverses into anxiety, which implies reversal from the paratelic back to the telic state.

Learning-drive mechanisms

(***) Because of its biological significance, it should be stressed here again that this learning mechanism is basically involuntary. In fact, this accounts for the "drive" for self-actualization in humanistic psychological theory. In the normal, healthy case, the telic and the paratelic state reverse sooner or later into each other automatically. And because of the open-ended character of paratelic behavior, the individual will time and again be launched willy-nilly into unexpected trouble. By seeking high arousal (excitement) whenever surplus-energy is available, experience is likely to be gathered involuntarily in situations of such gravity that they seem undesirable if foreseen.

Examples of how the above mentioned sequences of motivational changes occur, and indications of their importance for the processes of learning and development, may be derived from the work of various authors (Maslow, 1963, 1968, pp. 46-47; Ainsworth, 1977; Bowlby, 1969, 1977, 1979ab; van de Rijt-Plooij, 1986). They point out that growth occurs in small steps, and that each step ahead is facilitated by a feeling of security (telic). There is a necessity for 'points of reference' where one may seek refuge and rest every time when, after a long enough bout of exploration, the individual has, for the moment, reached his 'tax' of experience. This is clearly visible in the behavior of young children. For a young child which is, for instance, entering a new environment with its mother, it is characteristic to first cling to mother's knee while exploring the room with its eyes. After a while it moves away a little, while checking its mother's presence constantly. Subsequently, the excursions extend further and further away, while physical or eye contact with the mother is re-established time and again in between the (paratelic) explorations. In this way a child may explore and become acquainted with a dangerous and unknown world. If the mother suddenly disappeared, the child would grow timid and lose interest in reconnaisance of the world. It would merely be interested in getting back (telic), and may even lose command of already acquired skills---thus crawling instead of walking for example.

The child does not particularly 'strive' to enlarge experience and knowledge of the new setting. Rather, it expresses its curiosity in a balanced alternation with the need for safety and reassurance (Bowlby 1977, pp. 204-206). In such a way, the need to explore (paratelic) and the need for safety (telic) involuntarily help to bring about a proper sequence of experience and of opportunity to 'digest' such experience.

Bowlby empirically showed that for children it is of crucial importance to have opportunity to attain relaxation by being offered reassurance by a significant adult whenever they need it. If there is not enough reassuring support for the child and not enough opportunity for relaxation (no satisfaction of telic tendencies), the behavioural repertoire will not develop properly, and as a result neuroticism and other forms of behavioural imperfections will eventually develop (see e.g., Plooij 1979; Davenport 1979).

The same has experimentally been shown in chimpanzees. Van de Rijt-Plooij & Plooij (1986) and Plooij (1984) presented evidence that insufficient caretaking may cause a delay of many months for chimpanzee babies in taking their first steps. So, the peculiar picture emerges that a young chimpanzee walks earlier, if it has sufficient opportunity to hang-on to its mother's body. Whereas this may sound somewhat contradictory at face-value, it is quite plausible in the light of the theory developed above.

Chronic lack of reassuring (physical) support may not only lead to a delayed development, but may even lead to illness and eventually death Rijt & Plooij 1982.

On the other hand, maternal aggression and reluctance to respond to the infant's demands with reassurance and physical support, is not necessarily harmful. As van de Rijt-Plooij & Plooij (1986) point out, one of the essential tasks of the mothers at certain developmental stages is to actively "push" the infant over a threshold beyond which new behavioural systems may be triggered and exercised, leading to a next-higher level of maturation and autonomous functioning.

Whereas such a maternal "urging" may involve considerable amounts of aggression and some denial of previous support patterns, the behavioural development of young chimpanzees can be seriously impeded if their mother fails to urge them sufficiently to make them start performing different and for the infant novel behaviours.

In Selye's (1978) words, such maternal pressure and aggression and its emotional effect on the infant, geared to promote the learning process, could be labelled as "eu-stress". Selye argues that a certain frequency of stressful events is a prerequisite for optimal development (see Fig. 5).

Figure 5. Stress for Health (by kind permission of Arend van Dam)

Positive and negative learning spirals

(***) The Reversal Theory of antagonist motivational states may give us also more detailed clues as to the factors determining whether certain stressful events will in the long run lead to

  1. a cumulation of avoidance- and defence-reactions and to neuroticism and other behavioural deficiencies, or to
  2. a gradual mastering and digestion of the stressful experiences, leading to more integrated skills and abilities in that particular area.

It can in particular give us insight in the effect of given (physical) support and emotional security and reassurance on the processing of such stress-experiences. To that end we have to focus on the content of experience, rather than merely on its emotional flavour. An experience which is very frightening (anxiety producing) in a telic mood, may, because of its arousal raising properties, be attractive in a paratelic mood (i.e., intriguing). In particular, if there is ample opportunity for relaxation and recovery, more and more experience with the former source of anxiety may be facilitated in the successive paratelic states, until eventually enough experience with the situation in question has been gathered in order for this situation to have lost most of its arousal raising properties. By that time the situation has become either boring (paratelic) or reassuring because of its familiarity (telic), depending on the prevailing meta-motivational state. In other words, the situation has been 'mastered', and the formerly rough experience has been successfully 'digested'.

If, however, insufficient time and relaxation can be attained after every harsh experience, as to cause satiation of the telic state, the sequence as depicted in Figure 4, which is considered essential for the process of learning to occur in a satisfactory way, will be impeded. Instead of a 'final assimilation' of the experiences on a high level of integration, accumulation and fixation of defence- and avoidance-reactions on a rather simple, rough-and-ready level of functioning will then occur (see Figure 6) (see Peterfreund 1971 for more material on this effect).

In the case of optimum sequences of anxiety⇒relaxation⇒boredom⇒excitement⇒anxiety⇒relaxation⇒etc., experiences can be assimilated successfully and be transformed into highly integrated skills. And the increase in the variety of skills and in flexibility will then in turn facilitate telic/paratelic motivational reversals with sufficient time in the paratelic state, as is optimal for a further growth of skills and flexibility.

The other part of Figure 6 represents the opposite option. Both options are based on positive feed-back, the one resulting in a learning spiral with favourable results (a "positive learning spiral"), and the other one ending in an ever increasing accumulation of stereotyped reflexes and rigidity, with unfavourable consequences for further coping and learning (a "negative learning spiral").

The more situations and experiences of any kind have been experiences, re-experienced, and subsequently digested and masted, and thus have become familiar and maybe even reassuring, the easier it is to attain relaxation in any one problematic situation inducing the telic state. And this is the more likely if the previously mastered situations and settings are in some way related to that particular problematic situation in question.

Figure 6. Dependence of positive and negative learning spirals on the ability to establish proper rhythms of Telic/Paratelic alternations. Whether experiences are processed merely into stereotyped emergency-and avoidance-reflexes or into a high-level integration of skills and information, depends on the rhythm of emotional/motivational reversals: inability to reach relaxation⇒lower general skills level⇒lower ability to reach relaxation when needed⇒etc. (and vice versa).

Well-integrated experience and skills are most easily applicable in situations to which those experiences and skills bear some relevance. New skills and fields of mastery are therefore most likely to develop in areas of experience which are in some way related to other, already properly integrated and mastered areas of experience.

Skills therefore tend to grow in clusters, and conversely, unskills (sets of stereotyped avoidance reaction patterns) also tend to grow in clusters (see Figure 7).

Figure 7. Hypothetical example of the outcome of the growth of positive (+) and negative (-) systems of COndensed EXperience in the "field" of experience. Badly and superficially digested experiences (rendering stereotyped and rigid emergency reflexes) decrease the likelihood of a proper digestion of experiences in related areas of life, and vice versa. This accounts for Grof's (1976) positive and negative COEX-systems (systems of COndensed EXperience). Negative COEX-systems are e.g. fears, phobias, neuroses and consciousness blocks. This figure of course just represents a two-dimensional projection of the basically multidimensional space of all possible COEX-systems. Distances on the X and Y axes indicate a measure of unrelatedness of the experience.

Positive and negative clusters of COndensed EXperience

(***) A cluster of 'unskills' (see Figure 6) can sometimes be labelled as a "phobia" or as a specific form of 'neurosis', and indeed this is the case if the cluster of frightening items is—for the onlooker (e.g., a therapist)—easily recognizable as a 'special' setting. However, according to the view presented here, a behavioural repertoire which is just neurotic in general, may also be considered as an individual-specific set of phobias, resulting in a general restlessness, anxiety and unpreparedness for stress-evoking stimuli from the environment.

This prediction about the occurrence of clusters of related skills on the one hand, and of frightening and not-mastered items on the other, is supported and illustrated by the findings of the depth-psychological investigator and psycho-therapist Grof (1972, 1973, 1976). He analyzed some 2000 protocols of therapeutic L.S.D. sessions and concluded that for a more complete understanding of these sessions as well as of the personality structures involved, a new principle would have to be introduced into psychoanalytical thinking, which he called the principle of "specific memory constellations" or "COEX-systems" (systems of COndensed EXperience; see Table 2).

[...] A COEX system can be defined as a specific constellation of memories consisting of condensed experiences (and related fantasies) from different life periods of the individual. The memories belonging to a particular COEX system have a similar basic theme or contain similar elements and are associated with a strong emotional charge of the same quality. (Grof 1976, p. 46).
[...] The nature of these themes varies considerable from one COEX constellation to another [...] (Grof 1976, p. 47).

And he discussed for instance: systems connected with sex, systems that involve aggression and violence, systems related to humiliation and degradation damaging to the self-esteem, systems connected with guilt and moral failure, systems connected with emotional deprivation and rejection, etc.

[...] The personality structure usually contains a greater number of COEX systems. Their character, total number, extent, and intensity varies considerable from one individual to another. According to the basic quality of the emotional charge, we can differentiate negative COEX systems (condensing unpleasant emotional experiences) and positive COEX systems (condensing pleasant emotional experiences and positive aspects of an individual's past life). Although there are certain interdependencies and overlappings, separate COEX systems can function relatively autonomously. In a complicated interaction with the environment, the influence selectively the subject's perception of himself and of the world, his feeling and ideation, and even many somatic processes.[...] (ibid., pp. 47, 49).
Table 2. Negative and Positive COEX-Systems. Experiences which have previously been digested badly, and have thus been turned into a negative COEX-system (phobia, neurosis, etc.), may gradually be reprocessed and re-digested afterwards. This happens automatically if sufficient paratelic states do occur, since "frightening" issues may be considered "intriguing" in the paratelic state, precisely because they are arousal inducing.
COEX - systems (Systems of COndensed EXperience)
negative COEX-systems positive COEX-systems
flight/fight responses stay/play responses
behavioural rigidity flexibility of reactions
stereotyped reflexes creative responses
unskills skills
aversion from enjoyment of
phobia's interest in
neuroses mastery of
consciousness blocks
(unconscious repressions)
awareness of

Comparison of Table 2 with Table 1 shows that Grof's negative COEX-systems can be associated with a telic orientation whereas positive COEX-systems should rather be associated with a paratelic orientation.

One well-investigated and discussed example of a negative coex-system is Seligman's (see Kalma, this volume) "learned helplessness" syndrome. Seligman's experiments showed the incapacitating effect of the experiences inability of solving a stressful situation. The experimental setting enters into a negative COEX-system and every following experience of the sort is further consolidating a stereotyped and inadequate attitude in situations which are perceived as similar.

An illustration: the Neurotic Paradox explained

(***) The gradual successive growth of COEX systems by the mechanism of positive feedback—in the cybernetic sense—as described, could account for the latency of "incubation" period between the original traumatic events and actual neurotic or even psychotic breakdowns later in life. Manifest psychopathological symptoms seem to occur at a time when the COEX system reaches a certain critical extension, and traumatic repetitions contaminate important areas of the patient's life and interfere with the satisfaction of his basic needs.

It accounts also for what Mowrer (1950) indicated as the "neurotic paradox": the phenomenon that a basically adaptive mechanism, the acquirement of avoidance-responses, can be the basis for the acquisition of seeming maladaptive behaviour patterns, i.e., if a too strong and too continuous accumulation of such avoidance-patterns occurs. In fact, the above described mechanism (negative learning spirals) of the gradual growth of clusters of stereotyped avoidance reactions (negative COEX systems) accounts effectively for some frequently discussed empirical data that have been difficult to reconcile with classical conditioning theory, viz. the growth and perseverance of neurotic behavior patterns. I quote from Eysenck (1979, p. 158) (but see also Bindra's, Bolle's, and Öhmen & Ursins's commentaries on Eysenck, in Eysenck 1979):

In many neuroses we not only fail to observe the expected extinction of the unreinforced 'Conditioned Stimulus', but we find an incremental (enhancement) effect, such that the unreinforced Conditioned Stimulus actually produces more and more anxiety ('Conditioned Response') with each presentation of the Conditioned Stimulus.
[...] the final 'Conditioned Response' (the neurotic breakdown) is stronger (involves more anxiety) than the original 'UnConditioned Response'. This goes counter to all we know of the general fate of UnConditioned Responses; these are known to habituate, rather than to increase in strength[...] In [..] neuroses, traumatic original 'Unconditioned Stimuli' are distinctly rare (Lautsch 1971; Gourney & O'Connor 1971); in the majority of cases there is some sort of insidious onset, without any single event that could be called "traumatic" even by lenient standards (Rachman 1968; Marks 1969).

Whereas Eysenck takes great pains to squeeze these well-known clinical phenomena into a revised form of classical conditioning theory, which is basically a homeostatic theory (e.g., there is a tending for conditioned responses to return to the unconditioned strengths over time unless new conditioning occurs), the same phenomena may serve as perfect illustrations of the—non-homeostatic—theory of learning as developed in these pages, and of Grof's theory of COEX-systems. These characteristics of neuroses perfectly illustrate the negative options of these theories, viz. 'negative' learning spirals and 'negative' systems of COndensed EXperience, as well as the dynamic, non-static aspects thereof (see also Bolles' commentary in Eysenck, 1979). My model of motivational antagonists and learning processes predicts under which conditions a certain source of arousal (ambivalent stimulus) will enter into a 'negative' COEX-system with clusters of stereotyped avoidance reflexes, rather than to be 'digested' properly and to be integrated in a positive COEX-system with clusters of high-level skills. It all depends on the contingencies already present (innate, learned, or structured in the environment), and on the prevailing telic/paratelic motivational balance in the individual concerned (see also Bindra's commentary in Eysenck, 1979).

Mowrer's (1950) "neurotic paradox" is therefore no paradox at all, since it is based on a misconceived homeostatic character of the process of learning.

"Trying very hard" and the process of learning

(***) From the above it may be concluded that for acquiring skills in dealing with a certain setting it is generally speaking of much more importance that in and around that setting an adequate rhythm of telic and paratelic states is maintained, with sufficient paratelic bouts, than that one consciously "tries very hard" to master that specific setting in question.

Trying hard is typically a telic attitude. By trying very hard one may indeed gather relevant experiences in that setting. However, as this novel theory about the learning system explains, only by a proper sequence of telic and paratelic states with sufficient paratelic states the individual will in the end achieve "mastery" of the setting in question. It is especially in the paratelic bouts that the organism automatically and involuntarily by "curiosity" seeks out those experiences and those (sub-)settings that are best suited to fill out the "gaps" in the experience gathered up to that moment in time. That way the individual is enabled to automatically glue together the bits and pieces of experiences already gathered before. The telic / paratelic meta-motivational system has evolved to find the optimal balance between experiences that are without risk, but not very informative and experiences that are very relevant, but perhaps still too risky and difficult to handle. This motivational system steering curiosity in the most useful and fruitful direction operates involuntarily and unconsciously.

Trying very hard to acquire the desired experiences may therefore be counterproductive if its effect is a continuation of too frequent and long lasting bouts of telic states, blocking the emergence of sufficient paratelic meta-motivational states in between. The conclusion is that in any setting where a person seeks mastery, "trying very hard" should be applied very carefully. It may be useful to try one's best, but for optimal results there should be (created) enough space and time for paratelic states to emerge frequently enough. Curiosity, seeking its own direction, should have enough space to operate as it has evolved for.

One-sidedness of psychological theories

(***) Kalma (this volume) pointed out that in all major theories in the field of cognitive social psychology, the need for uncertainty-reduction is implied.

The present theory of learning, based on the notion of an antagonist Telic/Paratelic motivational and emotional system, clarifies, however, that all these theories primarily deal with the Telic aspects of behavior, and are therefore basically one-sided.

Kalma mentions amongst other things the "false consensus bias", the "actor-observer fundamental attribution error" (which means that the observer tends to disregard the effects of situations on their people's actions, focusing on their personal dispositions, while looking at themselves as reacting on situational events), the "cognitive dissonance" theory, "conformity defying available information", "ingroup-outgroup differentiation", dominance tendencies, dogmatism, rigidity, etc.. These concepts and behavioral models are all quite adequate description of the "reactive", telic part of the behavioral system, in which the individual seeks clarification, order and certainty in order to reduce arousal, even at the cost of a proper assessment of the situation. But all these theories fail to cover the paratelic side of our behavior.

"Strange means dangerous", but this only holds in the light of the telic state. In the paratelic state, "strange" means "exciting" which is then felt as pleasant and thus not avoided, but even sought!

The "actor-observer fundamental attribution error" is said to be stronger under stressful and ambiguous conditions. The present theory of learning states even more, namely that this cognitive behavioral effect is typical for the telic arousal-avoiding state, and that in the paratelic state there is no need for this sort of make-believe certainty, precisely because ambiguity, unpredictability of other people's behavior, as well as a feeling of not being in control of the situation, are all exciting.

Similarly, cognitive dissonance may be avoided at the cost of objective assessment when in the telic state, but in the paratelic state cognitive dissonance is rather associated with mirth or with evoked curiosity.

Hitherto, the paratelic, arousal-seeking side of behavior seems to be generally neglected in psychological theory. And since the paratelic states are the very basis of exploration, exercise and behavioral expansion, it is not very surprising that, similarly, psychological theories on "growth impulses" or on a behavioral expansive "learning drive", which forms the basis of humanistic psychological theory, are not too popular in scientific circles.

Evidently, this attitude is not warranted by the present theory (see van der Molen, 1985).

Behavioural idiosyncrasies and the dimension of "self-actualization"

(***) The organization of experience in positive COEX-systems and "skills", and and in negative COEX-systems and "un-skills", as predicted by the present theory of learning, has two important implications for the way people differ from one another.

On the one hand, the way the COEX-sytems come into existence and grow, readily explains the considerable behavioral idiosyncrasies among individuals. This learning system is basically open-ended, and consequently the available options of what is going to be learned are practically infinite. After all, this utterly flexible learning system is a recent and very special evolutionary achievement, providing a maximum potential for adaptation, a potential which is in particular characteristic for the human species.

On the other hand, as has been pointed out in Section 7 (see Figure 6), different "skills", no matter how unrelated they may seem (but, the more related, the better), enhance a further growth of skills in general, while unskills tend to hamper a further growth of skills.

The present theory therefore predicts that, apart from striking idiosyncrasies in the behavioral repertoires, individuals will tend to differ in some over-all level of skills. As a consequence, a dimension of 'general level of skills', or 'general rigidity versus flexibility', or 'general level of actualization of behavioral repertoire', or 'general growth level', or 'general level of exploratory and creative behavior', or whatever label one wishes to use, is likely to account for a substantial portion of the interindividual differences in behavior. And this runs essentially parallel to the basic idea in humanistic psychology of a dimension of 'self-actualization' being one of the most important sources of interindividual differences. (It should, however, be clear from the foregoing that according to the present theory of learning, no value-criterion can be attached to the notion of self-actualization, other than merely in terms of the smoothness with which the learning is running, and of the general level of skills acquired.)

Whereas it may be seen as a source of "basic injustice" that the already favored learn best and easiest, this mechanism is boosting biological selection. Differences in genetic make-up will also tend to become amplified phenotypically, and as a consequence the speed of selection is increased.

And such a booster effect on selective forces can in itself be regarded as another important asset of the mechanism of open-ended learning as paired to a reversal system of antagonistic motivational states, an asset which adds to its Evolutionary Stability.


This paper was written as an answer to questions put forward by Ian Mathie, University College, Cardiff, and Bill Livant, University of Regina, Canada, at the International Symposium on Reversal Theory, University of Wales, Sept. 2 - 4, 1983. Their stimulating remarks during that conference are gratefully acknowledged.

A modified version of this paper has been published in Acta Biotheoretica. The implications of this theory for psychotherapy and growth-psychology have been published in Apter, Fontana & Murgatroyd 1985.

(For returning to the "Guided Tour", click here)


Ainsworth, M.D.S. (1977), "Social development in the first year of life: maternal influences on infant-mother attachment", in Tanner, J.M., Developments in psychiatric research, London: Tavistock

Apter, M.J. (1976), "Some data inconsistent with the optimal arousal theory of motivation", Perceptual and Motor Skills (no. 43): 1209-1210

Apter, M.J. (1981), "Reversal theory: making sense of felt arousal", New Forum (no. 8): 27-30

Apter, M.J. (1982), The experience of motivation: a theory of psychological reversals, London: Academic Press

Apter, M.J. & K.C.P. Smith (1976a), "Humour and the theory of psychological reversals", in Chapman, A.J. & H.C. Fast, It's a funny thing: humour, Oxford: Pergamon Press

Apter, M.J. & K.C.P. Smith (1976b), Psychological reversals: some new perspectives on the family and family communication. Paper presented at the First International Family encounter, Mexico City.

Apter, M.J. & K.C.P. Smith (1976c), "Negativism in adolescence", The Counsellor (no. 23/24): 25-30

Apter, M.J. & K.C.P. Smith (1977), Religion and the theory of psychological reversals. Second Lancaster Colloquium on the Psychology of Religion, January 1977, Plater College, Oxford

Apter, M.J. & K.C.P. Smith (1979), "Sexual behaviour and the theory of psychological reversals.", in Cook, M. & G. Wilson, Love and Attraction, Oxford: Pergamon Press

Apter, M.J.; D. Fontana & S. Murgatroyd (1985), Reversal Theory: Applications and Developments, Cardiff, U.K.: University College Cardiff Press

Berlyne, D.E. (1960), Conflict, arousal and curiosity, New York: McGraw-Hill

Bowlby, J. (1969), Attachment and loss, vol. I, Attachment, New York: Basic Books

Bowlby, J. (1977), "The making and breaking of affectional bonds; expanded version of the fiftieth Maudsley Lecture, 19-11-1976", British Journal of Psychiatry (no. 130): 201-210 & 421-431

Bowlby, J. (1979a), "By ethology out of psychoanalysis: an experiment in interbreeding", Animal behaviour (no. 28): 649-656

Bowlby, J. (1979b), "Psychoanalysis as art and science", International Review of Psycho-Analysis (no. 6(3)): 1-14

Butler, R.A. (1953), "Discrimination learning by Rhesus monkeys to visual-exploration motivation", Journal of Comparative and Physiological Psychology (no. 46): 95-98

Davenport, R.K. (1979), "Some behavioral disturbances of great apes in captivity", in Hamburg, D.A. & E.R. McCown, The Great Apes, Perspectives on human evolution, vol. V, Menlo Park, Calif: Benjamin/Cummings

Dawkins, R. (1976), The Selfish Gene, Oxford: Oxford University Press

Eysenck, H.J. (1979), "The conditioning model of neurosis", The Behavioral and Brain Sciences (no. 2): 155-199

Eysenck, H.J. (1980), "The conditioning model of neurosis", The Behavioral and Brain Sciences (no. 3): 459-482

Fechner, G.T. (1873), Einige Ideen zur Schöpfungs- und Entwicklungsgeschichte der Organismen, Leipzig

Festinger, L. (1957), A Theory of Cognitive Dissonance, Evanston, III. - White Plants, N.Y.: Row, Peterson & Cy

Freud, S. (1888), "Hysterie", in Villaret, A., Handwörterbuch der gesamten Medizin, Stuttgart, at 886-892

Gourney, A.B. & P.J. O'Connor (1971), "Anxiety associated with flying", British Journal of Psychiatry (no. 119): 159-166

Grof, S. (1972), "Varieties of transpersonal experiences: observations from L.S.D. psychotherapy", Journal of Transpersonal Psychology (no. 4): 45-80

Grof, S. (1973), "Theoretical and empirical basis of transpersonal psychology and psychotherapy: observations from L.S.D. research", Journal of Transpersonal Psychology (no. 5): 15-54

Grof, S. (1976), Realms of the Human Unconscious. Observations from L.S.D. Research, New York: Dutton

Hull, C.L. (1943), Principles of Behavior, New York: Appleton-Century-Crofts

Lautsch, H. (1971), "Dental phobia", British Journal of Psychiatry (no. 119): 151-158

Lorenz, K. (1950), The comparative method in studying innate behaviour patterns. Symposia of the Society for Experimental Biology, Vol. 4, at 221-268

Lumsden, C.J. & E.O. Wilson (1981), Genes, Mind and Culture, Cambridge, Mass.: Harvard University Press

Marks, I.M. (1969), Fears and Phobias, London: Academic Press

Maslow, A.H. (1963), "The need to know and the fear of knowing", The Journal of General Psychology (no. 68): 111-125

Maslow, A.H. (1968), Toward a Psychology of Being, New York: Van Nostrand Reinhold (quotation from 1962 edition)

Molen, P.P. van der (1984), "Bi-stability of emotions and motivations: An evolutionary consequence of the open-ended capacity for learning", Acta Biotheoretica (no. 33): 227-251

Molen, P.P. van der (1985), "Learning, self-actualization and psychotherapy", in Apter, M.J.; D. Fontana & S. Murgatroyd, Reversal Theory: Applications and Developments, Cardiff, U.K.: University College Cardiff Press, at 103-116

Molen, P.P. van der (1986), "Reversal theory, learning and psychotherapy: On using the theory of motivational reversals for explaining the dynamics of psychological growth", British Journal of Guidance and Counselling

Mowrer, O.H. (1950), Learning Theory and Personality Dynamics, New York: Arnold Press

Peterfreund, E. (1971), Information, Systems and Psychoanalysis. An evolutionary biological approach to psychoanalytical theory, New York: International University Press

Plooij, F.X. (1984), The Behavioral Development of Free-living Chimpanzee Babies and Infants, Norwood, N.J.: Ablex Publishing Corp.

Rachman, S.J. (1968), Phobias: their nature and control, Springfield, III.: C.C. Thomas

Rijt, H.H.C. van de & F.X. Plooij (1982), Mother-infant relations in free-living chimpanzees of the Gombe National Park, Tanzania, Cambridge, U.K.: Unpubl. doctoral dissertation

Rijt, H.H.C. van de & F.X. Plooij (1986), The involvement of interactional processes and hierarchical systems control in the growing independence of early infancy, in Wind, J. & V. Reynolds, "Essays in Human Sociobiology", Study Series V.U.B. 2 (26): 155-165, Brussels

Schachter, S. & J.E. Singer (1962), "Cognitive, social and physiological determinants of emotional state", Psychological Review (no. 69): 379-399

Selye, H. (1978), "On the real benefits of Eustress (Interview)", Psychology Today, March issue: 60-70

Smith, K.C.P. & M.J. Apter (1975), A Theory of Psychological Reversals, Chippenham, U.K.: Picton Publ.

Zuckerman, M. (1974), "The sensation seeking motive", in Maher, B.A., Progress in Experimental Personality Research, vol. 7, New York - London: Academic Press