Maurice Guiraud-Rivière 1920
(1881 - ?)
monkey mirror

mise à jour du
4 mars 2004
1962; 196; 1258-1261
Behaviour of monkeys towards mirror-images
KR. Hall
Department of psychology, University Bristol


Mirrors have been used in field studios of birds to elicit various behaviour-patterns, and it has also long been known that anthropoid apes take, a great interest in the perception of their mirror-images. Thus Yerkes and Yerkes described how "in any captive chimpanzee, but especially in an isolated and, therefore, lonesome individual, the self-image, commands attention and arouses interest which loads to varied investigative expressions. In a word, the novel object is treated as if it were a fellow being. It may be touched repeatedly with fingers, hands, feet, lips, checks, or poked, slapped and struck gently or vigorously with hands, feet or both. Also there may be sudden reaches behind the mirror, if such is used as source of image, in seeming endeavour to surprise the other animal and grasp it. Social interest for a times waxes, as such behaviour runs its course; then it wanes, and presently the image is neglected".
Systematic observational or experimental studios of the behaviour of captive monkeys and apes towards their mirror-images or reflexions or to other mirrored objects or animals are, however, lacking. The present investigation arose as a result of preliminary observations of so-called 'imitation' and 'social facilitation' in four patas monkeys, Erythrocebus patas. Those monkeys had all been recently captured in Northern Nigeria, and wore obtained through the Wellcome Research Laboratories at Bockenham. Until they came to our own laboratories, they had not taken part in any kind of experimental situation. It is reasonable to assume that, in the wild, these animals would net see anything comparable to the mirror-image other than the reflexion of themselves on the horizontal plane such as is provided by stream or pond water. Hence, the initial, and subsequent, behaviour of such experimentally unsophisticated animals to mirrors or mirror-like situations seemed to be worthy of investigation.
chimpanze mirror
Of the four monkeys used in the work, two were full-grown adults, one male and one female, and two wore juveniles, also, one male and one famale. The exact age of none of them was ascertainable. They wore normally kept in separate, living cages, and, according to the exporimental situation, were net allowed to see or have contact with each other. The two young ones were kept in one room, and wore let out on most days to be together in a large cage which was divisible into two compartments according to the requirements of the social facilititation and imitation studios. The two adults were in a separate room adjacent to this, and they were also occasionally allowed out into the room. During the latter part of the observations, all four cages were, put in the same room while alterations to the other rooms wore being made, but appropriate screening of the cages was carried out when experimentally necessary.
Prior to any experimental observations being made, it was noted that whenever the two young monkeys were released into their living room, they showed intense and active interest in the images of themselves that they could see in the glass parts of the partitions dividing, on two walls, their room from the adjacent two rooms. The behaviour seen consisted of a variety of observing responses, such as standing on hindlegs, crouching low to the ground with head held low and staring, and of cautious approach towards the partition. In addition, the young female, in particular, frequently made 'gaping' or 'yawning' responses in which, while staring towards the reflexion, she lowered the head while opening the mouth to fullest extent, then raised the head again to the normal horizontal position. This reaction would occur with the animal in the sitting, normal standing, or low-crouched postures, and in partial form (for example, opening wide of the mouth without lowering of the head) as well as in the complets form described. The significance of this reaction, its variation in form and in circumtances of occurrence, and its relations to other behaviour components of the species'repertoire have yet to be fully worked out, so that the term 'yawning', to be used hereafter, is simply a usefully descriptive one. A similar reaction bas been described in wild chacma baboon males by Hall, and in captive rhesus monkeys by Rinde and Rowell. The interest to the two young animals of those reflexions in the glass partitions delayed any kind of social responses between the two often for several minutes after release from the cages.
The adult male's responses in the same kind of situation were even more prominent, the 'yawning' being often a reaction in which the head was lowered, when the animal was standing, almost to the ground. The adult female, on the contrary, tended to show cautious investigatory and observing responses and rarely, if ever, 'yawned'. This female had been a cage-mate, of the adult male for some wooks prior to coming to our laboratories. The adult male has shown a similar, but much briefer, kind of response on seeing his image in a pane of frosted glass standing upright against a wall.
Further indications of the value of this technique came from observing the behaviour of these animals when given access to a small hand-mirror in the cage or lying on the floor of the room. Here the monkey is able to move the mirror about, and the fascination of it to them was strikingly apparent. Indeed, its interest for the young ones in particular was such as frequently to counteract the attraction of preferred food items, such as fruit, in the vicinity of the mirror. Further, the two young animals tended to show as much interest in it when they wore together as when they were viewing it individually in their cages. In fact, the interest of one in it tended to be accentuated the investigatory and playful behaviour towards it of the other, so that social facilitation seemed here, as on many other occasions, to be operating strongly.
The typical behaviour of the two young when given the small hand-mirror separately in their cages, has consisted of a wide variety of investigatory behaviour, such as peering close up to it, licking at its surface, chewing the edge of it, pawing at its surface, picking it up and holding it in one hand or both hands, usually close up to the face, and sometimes putting one hand round to the back of the mirror as though trying to touch the reflected object. Both those animals have also frequently looked obliquely into the mirror, either when holding it up, or when it is lying flat on the ground, and on one occasion the young female 'yawned' while thus looking, and then immediately turned and 'yawned' at the observer whom she had apparently viewed in the mirror.
In a series of trials carried out over sevaral days, and using a variety of different surfaces of the same sized object ranging from completely black to full reflecting, the presentations being randomized to central for serial effects, there is clear evidence of differential stimulating effects according to the degree of mirror-image discernible. It is also clear, however, that there is a gradual habituation ta the mirrorimage as is shown by the fact that both young animals reduced their investigatory time with the full-reflecting hand-mirror front 100 par cent for each on the first 5-min trial to 60 and 63 par cent respectively on the fifth 5-min trial. There are, however, oscillations with some, of the intermediate mirror-surfaces which. suggest that the incomplete images may possibly be less subject to habituation than the complete, but these trends need further systematic investigation.
When the hand-mirror has boon placed on the floor of the room in which the two young animals have been freed from their cages, it still remains very compelling to them. Thus, the young female immediately on leaving her cage has come straight up to the mirror, peered into it as it lay on the floor, picked it up and held it close to her face, and carried it about with her, even though she was free to climb on the window ledges and look out or explore the rest of the surroundings. The young male also frequently attended closely to the mirror and manipulated it, and often has followed the reactions of the female to it. On one occasion he grabbed the mirror in his hands as she hold it to her face and tried to pull it away from her. She then ran away from him on her hindlegs clutching the mirror in both hands. This young female has always had priority of access to preferred food items, snob as fruit, when the two young ones have been together. However, when the rnirror has also been available to her, she has tended te leave only partly eaten such objects as bananas, and has even tolerated the young male eating a part of a banana very close to her while she has been gazing at her mirror-image.
The reactions of the adult animals, when the handmirror has first been placed in their individual cages, have been markedly different from those of the young ones. Thus, when it was placed flat in the food bowl in the adult male's cage, he reacted immediately with startle, and jumped away to the back of the cage. The reaction was evidently in response to the mirrorimage of himself because it occurred as soon as ho stood looking down from directly above it. He then avoided the mirror and went to sit at the back of his cage. Whenever he did glance towards it, his behaviour showed uneasiness in frequent 'yawning' and movements of the lips. It was only after 12 min from the time whon the mirror had been put in his cage that he again come and stood looking down on it, and again at once jumped away from it, 'yawned' six times in rather rapid succession, stood on his hindlegs above the mirror, and then gave a sories of thirteen 'yawns' at a rate of about one every 15 sec. He continued with this kind of behaviour throughout the whole period of the trial and also climbed around inside his cage very actively. Only after 15 min did ho for the first time put one hand out and touch the mirror very gingerly. During the next 6 min he picked up food pellets in his mouth, but without chewing and eating them, merely nibbling at their ends, and he also picked up nutshells that were lying near the mirror.
chimpanze mirror
The adult female's reactions ta the hand-mirror likewise showed startle, on first perceiving the mirrorimage, but then a cautious but more or less continuous investigatory behaviour shown chiefly through peering at the mirror, looking underneath it, twice putting her face right up against its surface, looking briefly into it, but never actually touching it with her hands. For the 5-min period following removal of the mirror from her cage, she continually searched around inside, the cage peering several times under the wooden board on the cage floor, looking under the food bowl in which the mirror had been lying, and so on. Thus, in a preliminary way, these observations on those four animals suggested the potential value of a more controlled study with systematic variations of the mirror-stimulus condition.
Tho first of these studios is described now in outline. A two-sided mirror, 30 in. high by 18 in. wide, was mounted vertically on a wooden faur-wheeled carrier, enabling it ta bc moved readily from. one part of the laboratory ta another. This mirror, sufficiently large for the animal to perceive the whole of its standing frontal image, was used on experimental trials at varying distances and conditions, while control trials were systematically interspersed with the experimental ones as a check on the effeetiveness of the mirror as such rather than of any other conditions surrounding the animals and affecting their behaviour. The control observations were made either when the mirror was set directly in front of the animal in exactly the same, way as for an experimental. trial except that the mirror-surface was kept covered with a white cloth, or when the mirror was exposed ta the direct view of the animal in the next cage but only to the indirect view of the central animal, or when the animal was screened from seeing anything but the usual laboratory environment. Common to all trials, experimental and central, were the presence of the same two observers, and the operation of recording equipment.
The behaviour of the two adult animals, whon confronted with the mirror separatoly and at distances of 3, 6 or 9 ft. from the front of the cage, indicated a high degree of tension in comparison with their behaviour in the central situations. Typically, the adult male stood with body tensed, or sat with his back straight and his head alert, staring into the mirror and 'yawning' repeatedly with the mouth gaping open wide and the head going low towards the ground. The head then tended to jerk back to the normal gazing position. On the first mirror trial with the male, this reaction occurred fiftean times during the first 60 sec, that is, once overy 4 sec. On this and subsequent trials on other days, the 'yawning' frequency decreased rapidly ta near zero a minute over the trial period of 10 min. However, as this response waned it gave way to other, and perhaps lower intensity, behaviour patterns, also indicative of tension. Thus, the male began repeatedly to pick up in his fingers the standard food pellets lying in a bowl in his cage and rubbed them repeatedly and for several seconds with a rolling motion between his two hands before nibbling the end of the pellet, without actually chewing any of it off, then finally letting the pellet drop to the floor and picking up another.
From observation of his normal undisturbed feeding behaviour with these pellets, it is clear that the rubbing is far in excess of that which he usually adopts with this food object, and it is also obvious that the nibbling and discarding of the pellets in this way is an unusual procedure for this animal. When engaging in this kind of 'false' feeding behaviour, the male sat sideways on to the front of the cage, so that he avoided seoing the mirror-image. In normal feeding, he tends to face towards the front of the cage.
A.nother indication of tension in the adult male's behaviour to the mirror seemed to be an increased frequency of self-scratching, as opposed to selfgrooming occurrences. Self-groorning is a leisuroly pickin-over of fur and skin, while self-scratching has here been a brief episode of rapid, dog-like scratching; for example, of the back of the neck by the hind foot. To what extent the mirror situation has affected the sexual behaviour of this animal is net yet clear; but it is possible that the observed occasions of erection of the penis and of masturbation during the experimental trials wore brought on by the conflict nature of the situation.
As a variation, the vertical mirror was once placed et a distance of 9 ft. from the front of the adult malo's cage, uncovered, and then moved slowly by the experimenter towards the cage. This elicited a dramatic and highly intense series of reactions. During the first minute, as the mirror was approaching him, he stood close to the front of the cage and 'yawned' 23 times, that is, at a frequency of 1 in about 2.5 sec. Thereafter, the 'yawn' frequency decreased sharply, giving a total for the 10-min trial of 67. When the rairror was right against the front of the cage, the cage door was slid aside se that there was no barrier between the male and the image. The effact of this was instantaneous, the animal et once striking repeatedly at the mirror with both hands, trying to bite at its surface, and pressing hard against it. This hand-scrabbling and pressing of the face towards the mirror, interspersed with occasional 'yawns', continued throughout the trial, except for brief intervals during which he scratched and sometimes picked up food pellets from the bowl but threw them immediately aside. When the mirror was moved away at the end of the trial, he continued to move restlessly about his cage, climbing around it, as though attempting to get out, and standing on hindlegs looking around the room. This approach trial was, as would be expected, much more effective in arousing intense and varied behaviour than the trials where the mirror was static.
The adult female, when shown the rnirror at a fixed distance in front of her cage, 'yawned' 14 times only in the first trial. This can be compared with the first trial frequency of 50 for the adult male. 'Yawning' was thus net the most prominent response of this animal to the mirror. Rather was her behaviour on the experimental trials characterized by cautions investigation, as shown by her low crouching postures, peering towards the mirror, looking continually around her cage, around the room, up te the ceiling and towards the cage of the adult male. Her agitation or uncertainty was, however, indicated by an enormous incresse, in frequency of self-scratching episodes during the experimental, so compared with the control, trials. On one trial, frequency of this response built up to a peak as her 'yawning' frequency waned, and 43 separate scratching opisodes were observed in the 10-mon trial, as compared with an average of 8 or less in control conditions.
This female's behaviour was markedly oriented much of the time towards the adult male on those experimental trials whore the screen between them was removed, enabling them to see each other. She repeatedly looked toward him, and he showed a high degree of alertness, standing repeatedly on hindlegs, gazing out in the direction in which she was looking, and occasionally 'yawning'. It seemed thus that he was responding, not to the 'common stimulus' provided by the mirror-image, for that was not visible to him, but to the agitated and watchful behaviour of his female partner. The mirror technique, used in this way on a pair of animals, seems to offer one simple way of studying in the laboratory what Miller, Murphy and Mirsky have called 'communication of affect'.
The two young anirnals, have only so far been given the vertical mirror when they are free together in their large living cage. The immediate affect of the mirror being placed in this cage was to inhibit their vigorous play-activity, including sexuel play such as mounting, biting, etc., and to concentrate the attention of both animals on the mirror itself. Therefore, far from being an attraction only for an isolated animal, as Yerkes and Yerkes have suggested, it has seemed to have an effect perhaps comparable with the first introduction a new monkey of about the same age as the two others. The young female, in particular, scarcely ever left the vicinity of the mirror while it was in the large cage, and, when she had peered into it, she scraped at its surface with her hands, and put her face right against it. She then continually walked round and round the mirror rather slowly and cautiously, looking at it from all angles. The young male, although also interested in the mirror, as indicated by his gazing full into it and waving his arms towards his own image, tended rather, as in other performances when the two have been together, to follow the female around, and hence to do more or less what she was doing.
The frequency data on the distinctive behaviour patterns, such as 'yawning', scratching, fumbling with food pellets, etc., may all, it seems, be used in full-scale experiments to indicate significant trends in the reactivity of the animals to the many variations made possible by the mirror technique. The study of habituation is itself, in the primates, a neglected aspect of learning. Our own preliminary data, using 'yawning' as the response selected for special attention because of its prominence, show a progressive waning in frequency over successive days, as well as within 10-min trials as from the first to the last minute. When more than one experimental trial has been carried out on the same day, the habituation effect has been even greater, and it is probable that it would turn out to be proportional to the time interval elapsing between the end of one trial and the start of the next. Two further points are suggested by the results. First, if an interval of 24 h occurs between the end of one experimental trial and the beginning of the next, there is a fairly marked initial increase in the frequency of the 'yawning' response. Secondly, as habituation of the te his mirror image progressed, se his more or less complets lack of response to the control situation of covered mirror or mirror te the side of his direct nview has tended to alter. Thus, on the fist control session, no 'yawns' were recorded, while 50 were recorded in the first experimental, session. On the fifth dy from the beginnig of experiments, the mirror elecited only 5 'yawns' four of them being in the first minute of the trial, and the control situation elecited 13 'yawns', but these were rather evenly spaced throughout the course of the trial. It would thus seem that, with the habituation to the mirror-image there is a conditioning of agonistic responses to the mirror-when-covered or to the situation in general, which therefore come to elicit a rather greater general level of tension than the normal These statistical trends are illustrated in Fig. 1 but it is clear that a complex of social and motivational factors must be operating in these animals which may affect all aspects of their behaviour towards the mirror-image. Thus, it is perfectly possible that the œstrous condition of the female may affect her behaviour towards it, as also may her proximity or otherwise to the adult male with whom she is on intimate relationships. Similarly, the reactions of the adult male to an apparent image to another adult male might be expected to vary considerably according to the social context.
Those preliminary observations suggest that the mirror technique, with its potentialities for experimental variation and for standardized presentation of 'social'stimuli, may be usefully developed to sample the behaviour of such animals in the restricted setting of laboratory space. Some of the learning processes involved in social approach and avoidance behaviour, highly complex in the natural group situation, can, it seems, with the aid of this technique, be selected for systematic experimental investigations which may help to bridge the gap between field and laboratory findings on non-human primate behaviour. At least in this way realistic situations can apparently be given to the animal so that they elicit a variety of conflict, frustration, and investigatory behaviour patterns, in circumstances where they can be objectively and quantitatively recorded.
Pandiculation: the comparative phenomenon of systematic stretching AF Fraser