Difference between revisions of "Striving, Playing and Learning"

Revision as of 21:08, 28 May 2012

On aggression and the dynamics of learning in the play and struggle called 'life'

By Popko P. van der Molen & Johan M.G. van der Dennen

Presented at the 1978 Congress of the "Ethologische Gesellschaft" at Basel and at the First Congress of the European Section of the International Society for the Research on Aggression (I.S.R.A.) at Strassburg, sept. 1981.

*** This page is under construction ***

Summary

A novel model on the dynamics of learning is presented which bears particular significance to the notion of (violent) aggressiveness. It suggests, among other things, that aggressive behavior cannot be properly understood whithout considering the basic dynamics of learning involved, without recognizing the shifting balance between 'striving and playing' and without recognizing the prevailing balance between social skills and 'unskills'. If any of these aspects are not considered, 'Aggression' or 'aggressiveness' will tend to be viewed as a separate phenomenon in its own right - which is unfortunately a widespread habit, not only in daily life, but also among the very investigators of the subject. The very use of labels such as 'Aggressiveness' and 'violence' keeps our attention trapped and hampers awareness of the underlying dynamic behavioral processes . In the first half of this paper some well-known models of aggression are critically reviewed and some of the basic concepts are subsequently integrated with theories of learning and with the 'reversal' hypothesis of antagonistic motivations, which replaces the optimum-arousal theory. An additional hypothesis on the long-term regulation of attention then completes the emerging model of learning. A model which stresses the antagonism between the possible growth of stereotyped and inflexible emergency reactions, and the long-term growth of varieties of subtle and flexible coping-response patterns (skills). The proposed model conflicts with many classical learning-conditioning concepts in that it predicts that the prevailing emotional (telic/paratelic, or centripetal/centrifugal) balance determines whether certain aversive and/or ambivalent stimuli result in the long run in attraction or in repulsion.

1. Critical recovery periods, an Experiment

In a series of experiments (Van der Molen, 1979, 198]), approximately 60 male housemice of the CPB's strain were weaned and isolated at the age of 21 days. Subsequently they were introduced every day or every other day into an unfamiliar observation cage, populated by a group of resident mice showing territorial aggressive behavior towards strangers. Being outnumbered by at least four to one and being on foreign territory, a trainee never had a chance to win and therefore was beaten up every time when introduced, and often slightly wounded. After 5 to 10 days of training, these trainees would immediately show avoidance behavior when being introduced, even when not under attack by any resident mouse at all.

There seemed, however, to be some sort of critical training frequency, i.e., if the periods of rest in isolation between the training introductions were too long, the effect of the whole training was the opposite of what was intended in the first place. If an 'experienced animal' or an experienced 'fleeing' trainee was introduced once every week or once every other week instead of every day, he would tend to put up skilled resistance. Such a male did not attack residents, but refrained from 'fleeing frantically' and appeared to 'stand firm' when necessary. By manipulating only the 'training frequency', trainees could be transformed from inexperienced males into either 'fearful fleeing males' or 'skilled fighters'. It should be remembered that neither type of trainee ever won a fight. During every introduction in their training period they had to fight against heavy odds. 'Skilled fighters' simply managed to hold out and fight back now and then without 'running for their lives' frantically. With the procedure described above trainees became either more skilled in fighting or more fearful than inexperienced males. Furthermore, in-experienced males varied more in their reaction patterns than did either of the two classes of trainees. Once a male had become a 'skilled fighter' it was very difficult to reshape him into a 'fearful fleeer' again. Apparently the same type of experience can lead to different learning processes, depending on the input frequency of the experience. The one direction in the learning spiral (deviation-amplifying mutual causal process: Maruyama, 1963) leads to stereotyped and intensive avoidance reactions; the other direction leads to increased skill in offering resistance.

Thus the temporal pattern of experience appears to be of great importance for its ultimate effect. Besides, individual mice differed in their critical digestion time for this type of experience; for the CPB's strain males in question it varied between 2 and 6 days. In general, males with a background of 'normal' social experience were rather well able to hold out against avalanches of territorial aggression. A 'normal' social history apparently provided ample opportunity for these mice to 'digest' their aggressive encounters in such a way that they became skilled in dealing adequately with agonistic social situations.

The following conclusions were drawn from these experiments:

(1) Male housemice can easily be trained in such a way that they perform a uniform type of avoidance behavior when introduced into a strange environment, inhabited by resident mice.

(2) The training required for turning an inexperienced male into a 'coward' is almost the same as the training needed to turn him into a 'hero'. The only difference is that in the former training schedule less 'digestion time' is available for every experience than in the latter.

(3) This 'critical digestion time' varies somewhat between individuals, and depends i.a. on their life-history.

1.1 Hormonal Feedback

A possible explanation of (part of) the phenomenon of 'critical frequencies of experience' may be found in the influence of fearful experiences on the endocrine system. Fear reduces for instance the testosterone level and with that probably the tendency to put up resistance when provoked (Bermond, 1977). Long training intervals allow this and other hormone levels to return to normal, but when the intervals between the fearful experiences are too short, the effects accumulate, thus rendering the individual inclined to flee instead of putting up resistance when challenged again. A considerable amount of experimental evidence is available to support this idea. For instance, Eleftheriou & Church (1968) demonstrated that in mice the blood plasma levels of LH, LH-RF and FSH decreased after aggressive encounters. The hormonal changes in losers appeared to be cumulative in contrast to the hormonal changes in winner-mice. Rose et al. (1972) demonstrated similar hormonal effects in rhesus monkeys. Many investigators (e.g. Bronson et al., 1973; Leshner, 1975; Schuurman, 1981, and in: Wiepkema & Van Hooff, 1977) found that aggressive encounters brought about stronger physiological changes (i.a., in blood sugar level) in losers than in winners. In more anthropomorphic terms: A conflict or encounter which elicits emergency reactions and raises anxiety in the subordinate may well represent an ordinary - eventually exciting - matter of routine to the dominant. All this means that apart from opposite conditioning effects of aggressive encounters on winners and losers on the neural level, a stable behavioral divergence between the social roles is still further enhanced and consolidated by different hormonal adaptations (Leshner, 1975). Similar effects have also been demonstrated in humans. Kreuz et al. (1972) conducted physio¬logical studies on trainees in a military camp and found that in periods of high tension, stress and extremely hard work, their testosterone level decreased considerably and recovered only gradually after relaxation.

1.2 Recovery Intervals and Learning

Apart from the hormonal feedback as depicted above, differential 'learning' processes on the neural level are involved as well (Leshner, 1975). Because animals in a Bèta-role and animals in an Alpha-role experience aggressive encounters in a different way (losing and winning resp.), different neural information is stored and processed. Besides, as e.g. Buchholtz (1978) and Trincker (1966) point out, there is a great functional and physiological difference between short-term memory storage and long-term memory storage. And there are limits to the amount of experiences of a certain type pro time unit that an individual can sucessfully digest and transform into (greater) long-term skills. This phenomenon has been established in a wide range of species from insects (Gross, 1978) to Man (Buss, 1973, pp. 278-280). The neural and hormonal feedback systems function synergistically. Whereas in lower developed species hormonal changes are relatively important, we may expect that in neurally higher evolved species neural mechanisms of adaptation will be relatively important in comparison with the purely hormonal mechanisms.

2. Some models of agonistic behaviour

Below, a general model is construed that covers adaptational mechanisms of various complexities and various functional levels, hormonal as well as neural, and including the higher learning processes in Man. In order to arrive at that model, we first proceed with reviewing some current models on aggression and on agonistic behaviour in general. For it is assumed that in particular those learning processes are of paramount importance for the individual, that lead to skills in dealing with agonistic encounters, other aspects of role-conflict, etc.

2.1 The Conflict- or Ambivalence Hypothesis

Somewhat oversimplified, this hypothesis states that agonistic and related behaviors are the product of activation of, and interactions between, two major antagonistic 'drives1 or 'tendencies': the attack- or approach-tendency and escape- or withdrawal-tendency. Many courting behaviors may similarly be considered as the product of activation and interactions of 3 tendencies, viz. the attack-, escape- and sex-tendency (see e.g., Tinbergen, 1952; Baerends et al., 1955; Kruijt, 1964; Hinde, 1966; Baerends, 1975). 'Approach' and 'withdrawal' behaviors may shade off into one another through intermediate - compromise - types of response, depending on the balance between the attack- and escape-tendency. Such compromise movements or postures are often postulated to be the evolutionary origin of a communicative signal or display. If the basic tendencies are highly activated, the antagonist behavior types tend to change in a more sudden and dramatic way than if they are moderately activated. In the former case attack may suddenly change into escape and escape may suddenly (catastrophically) change into attack.

This state of affairs may graphically be presented and mathematically described with the aid of the bistable models from the mathemathical branch of catastrophe theory as presented by e.g. Thorn & Zeeman (1974) and Zeeman (1976) (see Fig. 1). This model of agonistic behavior, however, accounts only for short-term changes in mood and behavior. No predictions are implied that reach beyond the particular agonistic setting under scope. The catharsis hypothesis, on the other hand, is primarily concerned with long-term predictions.

• • • Fig. 1 about here ***

2.2 The Psychohydraulic Model: the Catharsis Hypothesis

The catharsis hypothesis, when applied to aggression (Lorenz, 1950; Leyhausen, 1967; Eibl-Eibesfeldt, 1975), states that the longer an individual has not shown aggressive behavior, the easier aggressive behavior is evoked by external stimuli. Eventually the aggression-deprived individual will even start to seek opportunities to direct his aggressive behavior towards something or other and thus release his internally stored 'aggressive energy', ultimately to the point of releasing it 'in vacuo'. This process can be visualized through what has been called the 'psychohydraulic model', or, less respectfully, the toilet flush model.

Whereas this model may be quite adequate for describing some other functions like for example male sexual tendencies or feeding behavior, its use for describing aggressive behavior has become somewhat obsolete (Hinde, 1960, 1966; Manning, 1969; van Dijk, 1977; Zillman, 1979, van der Dennen, 1980; van der Molen, 1981); mainly because little evidence has been found that long-term deprivation of an opportunity to fight causes an increase in the readiness with which fighting can specifically be elicted. Yet appetitive behavior for agonistic interaction can nonetheless be demonstrated in a wide range of species: fish (Thompson, 1963; Rasa, 1971; Sevenster, 1973), birds (Thompson, 1964; Cole & Parker, 1971; Cherek et al., 1973), mice (Lagerspetz, 1964; Tellegen et al., 1969; Legrand, 1970; Tellegen & Horn, 1972; Connor, 1974; Kelsey & Cassidy, 1976; Connor & Watson, 1977), rats (Dreyer & Church, 1970; Taylor, 1975), hamsters (Eibl-Eibesfeldt, 1971), and monkeys (Azrin et al., 1965).

But in every case investigated, such alleged appetence for aggressive behavior could equally well be labeled as appetence for certain other functional pattern such as e.g. territorial behavior and/or as attempts of the individual to maintain an optimum level of arousal. "Excitement and kinetic activity are shown to be dependent on the external stimulus state of the animals decreasing under conditions of low mean environmental stimulation. These findings indicate that an animal attempts to regulate its internal stimulus state by behavioural means when the component variables of this state have been disturbed by environmental conditions" (Rasa, 1971).

Sevenster (pers. comm., 1980) arrived at similar conclusions in his experiments with sticklebacks, when evaluating his results on the rewarding effect of opportunities to interact aggressively. He investigated the rewarding properties of each component of the situation of agonistic interaction in question separately, and concluded that it is, in particular, the element of environmental change which works rewarding for the (somewhat arousal-deprived) male stickleback, but only so, if the change in question induces no flight behavior. Parallel to these findings are experimental results as those of Kavanau (1967), who showed poignantly how mice, being forced to respond in stereotyped experimental situations, will give 'incorrect' responses as a means of introducing variability.

From all these considerations we may conclude that for understanding function and (long-term) causation of aggressive behavior it is apparently of great importance to include phenomena such as 'boredom', 'excitement- seeking' and 'anxiety' in our models.

Specific appetence for aggressive behavior may in fact not exist at all in ordinary natural settings in which plenty of opportunity is available for attaining proper arousal levels. This is also suggested by, for instance, Goodall's (1971) descriptions of the occurrence of aggressive encounters in the daily life of free living chimpanzees. The frequency of intraspecific aggressive behavior varies strongly for every individual and for the whole group in question, and is highest whenever hierarchical relations have become unclear. The readiness to show intraspecific aggres¬sive behavior does not seem to depend at all on the period of time that has elapsed since the last agonistic confrontation, but rather on circumstances which make it difficult to avoid the application of coercive means.

The existence of a drive for aggression, irrespective of a functional context, would, in fact, be highly improbable from an evolutionary point of view (Craig, 1918, 1928; Tinbergen, 1956; Hinde, 1960, 1970, 1974; Marler & Hamilton, 1966; Wilson, 1971, 1975; Scott, 1973; Crook, 1973; Schuster, 1978), the rationale of which has been analyzed in game-theoretical and 'Evolutionary Stable Strategy' terms (Hamilton, 1971; Maynard- Smith, 1974, 1978; Maynard-Smith & Price, 1973; Parker, 1974; see also: Wilson, 1975; Dawkins, 1976; Fry, 1980). An individual runs heavy risks every time he engages in aggressive encounters. Therefore it seems good strategy to reserve aggressive behavior for situations in which the risk of damage is sufficiently counterbalanced by a possible raise in the chances of survival and/or propagation of the specific genetic information in question, after successful agonistic action.

It would appear that the regularity with which the males of certain species carry out charging displays can be more adequately explained by the 'hierarchico-cybernetic model' with the aggression-system as a low- level subroutine (Baerends, I960; Tinbergen, 1969; van Dijk, 1977) than by an energy model of aggressive motivation (Lorenz, 1963; Rasa, 1971, 1980).

2.3 Aggression and Emergency Behavior

Van Rooijen (1976) reviewed models of 'escape-' and 'emergency-' reactions as derived from current ethological and psychological research. In his study he described a general 'emergency-system' which prepares individuals for responding adequately in cases of emergency, and he labels it 'Anticipation of Damage System' (ADS). An essential of van Rooijen's model is the application of a strategy of 'flexibel response'. The more an individual is threatened, the stronger his ADS is activated. The level of activation determines which types of behavior can be expected. Furthermore, individuals can 'learn' which situations are best met with a careful approach and which situations require instantaneous flight or attack. Van Rooijen suggests that the learning processes themselves are probably enhanced most by a moderate activation level of the emergency system. Considerable information uptake- and processing capacity is required for learning, whereas in the most extreme emergency-reaction-modes such as e.g. panicky flight, the full neural processing capacity is probably reserved for ad hoc problem-solving. When the ADS is activated too strongly and/or too continuously, pathogenetic effects emerge, such as neuroses, phobias, psychosomatoses, and even physical disorders (Blythe, 1973). These pathological traits in their turn enhance, through inadequately rigid responses in an ever-varying environment, strong stimulation of the emergency system. Thus vicious spirals ('Teufelskreize') of ever increasing pathogenic effects are likely to occur as long as an individual does not often enough and long enough succeed in reducing the alarming stimuli and the resulting level of arousal.

Agonistic behaviors may be fitted into van Rooijen's model in two ways. On the one hand, flight and attack may be final - suddenly and catastrophically occurring - reflexes during high activation of the emergency system. On the other hand, flight and, in particular, attack can be learned and may in fact be optional responses to specific familiar emergency situations. In that case flight and attack may already occur at a moderate activation level of the emergency system, and consequently tend to change more gradually into each other.

2.4 Aggression,and Behavioral Reduction versus Behavioral Expansion

Kortmulder's (1974) hypothesis of 'Behavioral Expansion' runs to some extent parallel to van Rooijen's ideas. To the classical postulate of antagonistic attack- and escape-tendencies, Kortmulder adds the antagonism between 'centrifugal' and 'centripetal' tendencies. Full-scale attack and flight are both considered to be centripetal behaviors whereas courtship and, in particular, playful (agonistic) interactions are considered to be more centrifugal types of behavior. The terms centrifugal and centripetal refer to the tendency of expanding the (interactive) behavioral activities, respectively to the tendency of reducing and simplifying the (interactive) behavior situation. Strong activation of the 'emergency system' (van Rooijen) corresponds with 'centripetal tendencies' (Kortmulder), and a weak activation of the 'emergency system' corresponds with 'centrifugal tendencies' of behavioral expansion.

When exploring the environment or when playfully interacting with a partner, an individual is supposed to be in a process of 'behavioral expansion' (centrifugal behavioral forces; cf. also Orme-Johnson's, 1973, 'Stay-and-play response'). But when the situation becomes dangerous - for instance when a partner suddenly play-attacks too frighteningly hard - the 'Anticipation of Damage System' is activated and an increased centripetal tendency leads to reduction of the behavioral setting through flight or attack.

According to Kortmulder, individuals strive after being in a state of behavioral expansion as often as possible. Through that behavioral state their behavioral repertoire expands and their skill in dealing with the animate and inanimate environment increases through experience.

Parallel to Maslow's (1968) behavioral models of men and other primates, Kortmulder (1972, 1974) discerns between (a) animals with a behavioral repertoire which is fully actualized through sufficient periods of behavioral expansion (self-actualizers), and (b) animals who have stayed under the influence of centripetal behavioral forces too often and too continuously and who subsequently show a reduced, 'truncated', and highly stereotyped behavioral repertoire. "Whereas self-actualizing fishes typically swim around actively also outside reproductive periods, or feed, inferior fishes often stand still for long periods of time, even in the absence of the dominant male. If they move, locomotion is typically, wavering and slow. They often do not feed at all ........ The general behaviour of such inferior fishes is somewhat similar to the behaviour of a sick fish. This is so even in the absence of any visible trace of bodily damage. The above observations suggest that the switch from self-actualizing to inferior behaviour marks the onset of a profound physiological change which is easily reversible when short-lasting but which tends towards irreversibility when submission is prolonged" (Kortmulder, 1972).

== 2.5 Butterfly Catastrophes ==

Taking the models of Kortmulder and van Rooijen into account, we can explain not only the gradual and the sudden changes from fight to escape and vice versa, but also the gradual and sudden changes from play to full-scale agonistic interaction. Thus we may integrate in one model the two different modes of aggressive interaction, viz. the 'rough- and-tumble' and the 'reactive fighting' as may be discerned in many species (rhesus monkeys: Harlow, 1965; chimpanzees: Goodall, 1971; human children: Blurton-Jones, 1967; Hamburg, 1971).

Like the mechanisms as postulated by the Conflict- or Ambivalence Hypothesis (see fig. 1), the mechanisms as described above may also be represented graphically (and mathematically) by means of a model from catastrophe-theory. Butterfly catastrophes (Zeeman, 1976) describe tri- stable systems, for instance three such behavioral modes which can shade off into one another, but which may, under certain conditions, also turn into each other in a sudden and catastrophic way (see fig. 2).

• • • • • • • • • • Fig. 2 about here ***********

One of the parameters of the graph in fig. 2 is the so-called 'butterfly-factor' which determines the size of the pocket in the bifurcation-set. A bigger pocket represents a greater probability of playful and explorative behavior. The 'butterfly-factor' might therefore be used to indicate an individual's skill in dealing with ambivalent motivational situations without having to resort to full-scale attack or flight. In other words: by an increasing butterfly-factor we may graphically represent the long-term effect of adequate frequencies of behavioral expansion, namely the raise in skills which results from the centrifugal behavioral activities of play and exploration.

A raise in skills causes in its turn a lower necessity to resort to centripetal behavior and thus a lower .probability of purely agonistic behavior and consequently a further increased probability of playful and explorative behavior.

This positive feedback mechanism in learning represents the opposite of the vicious spirals of ever-increasing pathogenic effects mentioned above.

Of course, it is also possible to block the positive learning-spiral by denying the individual any reasonable amount of experience, for instance by isolation. Centrifugal moods cannot then simply result in drifting into relevant experiences through lack of environmental possibilities. Individuals which have been isolated during long and crucial periods of their lives, generally exhibit inadequate social reactions (notably, anomalous sexual and agonistic behavior) to a high degree (jungle fowl: Kruijt, 1964; mice: van der Molen, 1979; rats: Peijs, 1977; rhesus monkeys: Harlow, 1961, Nowak & Harlow, 1975; squirrel monkeys: Hopf, 1981). As Kortmulder (1974) points out, courtship may be compared to play in that courtship partners 'learn' to get behaviorally adjusted to, and synchronized with, each other and thus 'learn' to avoid full-scale agonistic behavior towards one another. In the course of the courting process centripetal tendencies gradually give way to centrifugal tendencies until mutual aggression has virtually disappeared and the animals have learned to 'know' each other in playful interaction. Therefore Zeeman's graphical representation might 'mutatis mutandis' be used equally well to describe the interactions between sexual-, approach-, withdrawal-, attack- and escape-tendencies in courtship and the learning processes involved.