Effect of brain asymmetry on learning and feeding in chickens.

Abstract

EFFECT OF BRAIN ASYMMETRY ON LEARNING AND FEEDING IN CHICKENS LESLEY J. ROGERS SUMMARY This paper suggests that asymmetry of brain function may have important implications for display patterns and social organisation in chickens. The functions known to be asymmetrically organised in the chicken brain are discussed together with the techniques which have been used to reveal these asymmetries. Hemispheric and eye-eye asymmetries have been demonstrated. The latter are manifest in behaviour of the whole animal, such that attack and copulation responses are more likely to be elicited by stimuli in the left visual field, and the right eye is better in performing visual discrimination learning to feed. Suggestions are made for management relating to the use of polypeepers and position of feeders in the cage. The effects of domestication, incubation conditions and hormonal treatment on asymmetry are discussed. I0 INTRODUCTION The left and right hemispheres of the chicken brain have been shown to have different functions in the control of many behaviours. Such asymmetry of brain function has now been reported for a number of species (Harnard et al. 1977; Denenberg 19811, but in avian species it has special relevance to the behaviour of the whole animal. The reason for this is that the optic completely in birds, and birds have no major nerve fibres decussate interhemispheric connecting system such as the corpus callosum of mammals. Thus information received by one eye is processed largely by contralateral side of the brain (Rogers 1985). In other words, information processing by the left-eye-system, LES, differs from that of the right-eye-system, RES, and therefore the left and right eyes look at different worlds and direct That is, in birds, asymmetry of brain different behavioural sequences. function becomes manifest at the perceptual level and, for the visual system at least, it becomes an asymmetrical template imposed upon the space surrounding the animal. As a consequence of this, asymmetry of brain function may have become manifest in social behaviour, and so play a role in the management of commercial chicken flocks. II . ASYMMETRY IN THE CHICKEN BRAIN Asymmetry of function in the chicken brain (australorp x leghorn) was' first demonstrated by unilateral administration of cycloheximide (CXM), an inhibitor of ribosomal protein synthesis, into either the right or left forebrain hemisphere on day 2 of post-hatching life (Rogers and Anson 1979). Treatment alone was found to retard of both hemispheres or of the left hemisphere learning of a visual discrimination task, in which the chicken has to search for grains of food mash scattered on a background of small pebbles (Rogers et Treatment of the ,right hemisphere with CXM was found to have no al. 19741, learnt as well as effect on this learning performance; these birds The learning deficit which follows treatment of the saline-treated controls. left hemisphere can be detected long after the CXM has ceased to inhibit protein synthesis. Learning performance is usually measured in the second Physiology Department , 2351, Australia. University of New England, Armidale, New South Wales week of life, but the deficits have been found to persist up to 20 weeks at The CXM treatment in early life must disrupt least (Rogers et al. 1974). brain development such that subsequent information processing is impaired. This disruption is now known to occur, not directly by CXM*s inhibition of protein synthesis,, but indirectly by its effect in causing the temporary accumulation of the amino acids glutamate and aspartate in brain amino acid These amino acids, which are neuronal pools (Hambley and Rogers 1979). excitotoxins, cause the same effects as CXM when they are administered alone to the left hemisphere (Rogers and Hambley 1982). As retardation of visual discrimination learning occurs after treatment of the left hemisphere irrespective of whether the right hemisphere has been treated or not, the simplest hypothesis to explain these results is that the pathways involved in learning to discriminate grain from pebbles are located in the left hemisphere. This asymmetry can also be demonstrated by testing untreated chicks monocularly on the same task. If untreated male chickens in their second week of life are tested monocularly on the visual discrimination task, those with occlusion of the left eye, and therefore using their right eye and left hemisphere (RES), learn significantly more slowly than those with occlusion of the right eye, and therefore using their left eye and right hemisphere (LES) This asymmetry is consistent with that (Andrew et al. 1982; Rogers 1985). However, while it is revealed by intracranial administration of drugs. present in monocularly tested males, females chickens tested similarly in the second week of life show no eye-eye asymmetry for this task (Rogers 1985; Given that females do show the asymmetry revealed Andrew and Brennan 1984). by unilateral drug treatment of the hemispheres, this suggests sex differences in the organisation of visual pathways from eye to brain in the chicken. Rogers and Anson (1979) found that, irrespective of whether they learnt to discriminate grain from pebbles in the task, chicks tested using the RES in general pecked more rapidly than those tested using the LES. Consequently, they proposed that the right-eye-left-hemisphere system may drive pecking to feed. In a different task which examined pecking rates at a novel, bead stimulus, Andrew et al. (1982) found that pecking performance of the RES was the same as that of birds tested using their binocular field, but This led them to deducte that right-eye that of the LES was different. The RES has mechanisms dominate control of pecking in the binocular field. also been found to dominate for recall of memory of an operant, visual All three of discrimination task for food reward (Gaston and Gaston 1984). these paradigms implicate the right-eye-left hemisphere system as dominant for control of pecking and feeding behaviour. Chickens treated with CXM or glutamate in both hemispheres, or the left hemisphere only, also show slowed auditory habituation learning and attentional pertsistence (Anson and Rogers, 1979; Howard et al. 1980; Rogers, Treatment of the right hemisphere has no effect on these behaviours. 1985). An attentionally persistent animal switches attention from one stimulus type to another less frequently (Andrew and Rogers 1972; Rogers 198% Attentional persistence has been' studied on visual tasks for choice of food grains. Chicks treated with drug in the left hemisphere or both hemispheres switch Since search from red to yellow coloured food grains less frequently. attentional persistence is manifest in the visual system (it may be present in other modalities as wellL it is likely to be present as an eye-eye asymmetry with implications for feeding and social behaviour, It should be noted that attentional persistence is distinct from retarded learning, since the two behaviours possible to be manifest known about can say as the It is not yet be clearly separated (Rogers and Anson 1978). whether the asymmetries in auditory habituation are likely to asymmetries in behaviour of the whole animal9 because less is organisation of the auditory pathways from receptor to brain. The forebrain also plays an asymmetrical role in the control of Treatment of the left attack and copulation behaviour in the chicken. hemisphere of male or female ckicks with CXM or glutamate leads to elevation of attack and copulation scores, just as if the chicken had received treatment with an androgenic steroid (Howard et al., 1980). No such elevation of these behaviours occurs after treatment of both hemispheres or of just the right. Given that treatment of both hemispheres has no effect in this case and elevated scores only occur after treatment of the left, the left hemisphere must know the state of the right, and vice versa. In this respect asymmetry for attack and copulation differs from visual discrimination learning as revealed by treatment on day 2 of life. Asymmetry for control of attack and copulation is present, but it requires interhemispheric coupling. The asymmetrical control of copulation is also evident as asymmetry between the eyes. If birds treated with testosterone are made monocular from day 5, before copulation testing commences, those using their LES show elevated scores equivalent to the levels reached by binocularly tested birds, but those using their RES do not (Rogers et al. This occurs in both male and female chicks, although different steroid doses are needed to reveal These data suggest that the right the effect in each sex (Bullock 1985). hemisphere activates copulation while the left hemisphere suppresses it tests since (Attack could not be accurately investigated in monocular monocular chicks show variable attack scores depending on whether the chicken has caught sight of the stimulus or not. However, the asymmetry for attack was similar to that obtained for copulation.) 1989 l Comprehensive accounts of the asymmetry present in the chicken brain To summarise have been given in two reviews (Rogers 1980; Rogers 198% briefly, the right eye-left hemisphere system contains the pathways necessary for visual discrimination learning, it dominates control of pecking in the binocular field, is involved with switching attention from one stimulus type All to another, and it suppresses the performance of attack and copulation. of the functions of the RES taken together, indicate that it may have a dominant role in feeding behaviour, involving choice of food type and search for food. The left eye-right hemisphere system activates attack and copulation, and, as Andrew and Brennan (1983) have shown, it is also the system which has increased responsiveness to novelty and fear. Andrew (1982) has therefore suggested that the right hemisphere is active during the intense performance of species-specific responses, and he suggests a parallel to the right hemisphere's involvement in emotional responses in humans. III. CONSEQUENCES OF ASYMMETRY ON BEHAVIOUR OF THE CHICKEN As a consequence of the asymmetrical organisation of the brain and the completely decussating eye-to-brain visual pathways, the position of a stimulus in the visual space surrounding a bird would be expected to influence not only the likelihood of respon:se to that stimulus but also the nature of the response. Stimuli received in the left visual field may be more likely to elicit attack or copulation responses. Of course, the bird may well overome any such initial bias on responding by always turning its head to use both fields of vision, at least to use t;he small area of binocular overlap, before it responds. If so, this should be e*trident in the head movements made before No measurements of these have yet been made, It remains quite responding. possible that in situations requiring rapid responses9 possibly attack pecks9 there is a left-side bias for increased likelihood of a response being elicited. When chickens must discriminate fine differences between grains or seeds and the background they usually use their binocular, frontal field of vision. Although the RES dominates the biocular field for pecking to feed, a side-bias in discriminating grain and pecking would only be evident when perpheral vision is used, such as when grains pecked at bounce either left or right. Performance of this kind should be best is the right visual field. Certainly, when a chicken is feeding and discrimating grain or seed in the frontal field, it would be much more likely to be distracted by stimuli, possibly predators or other birds, moving into the left peripheral field of vision, rather than the right. The response to novelty in the left visual field is greatest; thus feeding is more likely to be distrupted by a response of fear, attack or even copulation. Recognition of these side-biases may be important for care of commercially raised chickens, If, for example9 the food trough is placed so that new animals approaching it are likely to approach a feeding animal from the latter's left side* more disruption of feeding, and possibly increased aggression, may occur. The order of a group considerable amount to do with disrupts such orders may increase of these predictions are directly standing at the feeding dish may have a these side-biases. Movement of feeders which All aggression and disrupt the peck-order. testable. In the case of copulation, the directing of circling of the cock during courtship waltzing may determine whether mounting will occur or not (a greater chance of mounting if circling is in the anticlockwise direction). Alternatively, the cock may choose to waltz in one direction or the other depending on its motivational state9 whether it intends to copulate or not. Either way, the visual field asymmetry would be incorporated into the courtship display (Rogers 1985). flocks of Polypeepers are used to reduce aggression in large intensively reared chickens. They would seem to achieve this by occlusion of the binocular field of vision. However9 it may be necessary to occlude only the left eye in order to reduce aggression, This is provided, of course, that occlusion of the left binocular field only will suffice, a factor which needs to be tested since the experiments to date have used monocular occlusion of the complete visual field on one or the other side. The occlusion of the left eye should also occur early enough. In the experiments demonstrating LES-RES asymmetry for attack and copulation the monocular occlusion was made by placing the eye patches over the ejres before testing commenced and these patches remained in place throughout the next 10 days or so of testing. With this paradigm, chicks using the LES showed elevated scores but those using their RES remained low (at control levels) throughout the entire testing period. In another experiment the chicks were tested binocularly for attack and copulation until the second week of life. The eye-patches were then applied, at which time the LES performance? dropped to base-line levels while the RES remained high, as predicted (Zappia & Rogers, 1985). However, this asymmetry was only temporary; subsequently*, the RES scores increased to those of the this reason, it would seem necessary to put the LES. For monocular-polypeeper on the chicken before attack and copulation performance under the influence of testosterone's action. increase In the levels experiments reported here juvenile attack and copulatory behaviours were stimulated in early life by testosterone treatment. In the normal course of hormonal and behavioural changes do not occur until development these This would be the age at which to apply the monocular occluder. puberty. Application of a monocular polypeeper to the left eye would not only be expected to be sufficient to reduce aggression, but it would leave an unobstructed RES available for discrimination of grain for feeding. The polypeepers which occlude the binocular field of both eyes must either cause deficits in visual discrimination of food grains, since unobstructed visual discrimination learning primarily uses the binocular field, or it must require behavioural adjustments to overcome the loss of visual field. The monocular, occluder of the left eye may obviate any need for such adjustments. IV . ASYMMETRY AND DOMESTICATION Preliminary evidence 'suggests that asymmetry of brain function is present in the feral strain collected from North West Island, and now breeding at the University of New England. Asymmetry of brain function in the chicken has therefore not simply been selected for by domestication. (a) Methods Eggs of the feral chickens were incubated in dark, force draft incubators until day 19 of incubation, at which time they were placed into a lighted incubator until hatching. After hatching they were held in groups of 4 for the first 3 days of life. Half of the animals were injected with glutamate (5 L 100 mM) into the left hemisphere and 5 L of 0.9% pyrogen-free The other half received glutamate in the saline into the right hemisphere. right hemisphere and saline in the left. On day 3, each chicken was placed in an individual cage by 12 cm high) which isolated it visually, but not auditorally, chickens. Testing for attack and copulation commenced on day These behaviours were scored continued daily until day 15. hand-thrust tests (Zappia and Rogers 1983). (9 cm square from the other 7 of life and using standard For attack testing the hand is thrust with arched fingers at the Responses range from avert gaze to active level of the chicken's beak. sparring and attack leaping, and the score is made according to a ranking For copulation testing the flattened hand is thrust at order from 0 to 10. Responses range the level of the chicken's chest and then held stationary. from no mounting of the hand (zero score) to mounting the hand with crouching, treading and pelvic thrusting (maximum score of 10). Each of these tests were repeated three times daily and a mean score taken. (b) Results After being isolated, 70% of feral chicks treated with glutamate in the left hemisphere failed to eat enough to maintain a normal rate of body growth rate, and possibly also to drink enough. Many of these animals died By comparison only 30% of the chicks before testing commenced on day 7. treated with glutamate in the right hemisphere showed the same effects. It should be noted that no such lack of ability to maintain body weight is evident in the australorp x leghorn strain. In this preliminary experiment only 6 feral chicks (male and female) remained in each group to be tested for attack and copulation. Unlike the previous findings for australorp x leghorn chickens, there was no difference in attack performance between the groups of feral chicks treated in the left or right hemispheres. Both groups scored very high levels of attack, a result which we have also found in untreated feral chickens. However, the group treated with glutamate in the left hemisphere showed elevated copulation scores while that treated in the right did not (e.g. P = 0.002 for a l-tailed U-test comparison of the scores on day 15). (c) Discussion The same asymmetry for control of copulation found previously in the australorp x leghorn strain is present in the feral strain. (Note that the intracranial injection technique reveals asymmetry in both males and females of the commercial strain, and we also found no sex difference in the feral strain when using this technique). As attack scores were so high in all the feral chickens no 'asymmetry for this behaviour could be revealed using a Perhaps asymmetry for pharmacological method which unmasks attack behaviour. control of attack can be demonstrated by monocular testing, but, as mentioned earlier, this too has problems for attack scoring. The high number of feral chicks which failed to feed adequately after glutamate treatment of the left hemisphere is interesting given previous suggestion from experiments with commercial strains that it is the left The effect in ferals is more hemisphere which directs pecking to feed. It is also the left hemisphere which processes learning to visually dramatic. discriminate food grains, and these data for feral chicks may suggest that disruption of its development by glutamate can become life threatening. Further experiments need to be conducted using the feral strain, but it can already be seen that asymmetry of brain function is present in this strain. V`, FACTORS AFFECTING ASYMMETRY While domestication has apparently n0.t s,e;lected for the presence of brain asymmetry in the chicken, incubation conditions can have significant Ro.gers and Anson (1979) suggested that the asymmetry of brain effects on it. function in chickens may. at Ieast in part, be generated by unequal growth of the left and right hemispheres, which occurs as a result of unequal light input into the left and I-jrght eyes befo,re hatching. In the last days of incubation, whe-n the visual pathways from eye to brain are making functional connections, the chick: embryo is oriented in the' egg such that the body and wing occlu-de the left eye, leaving the right eye exposed to light enter5ng through the egg shell and membranes. The greater amount of light received by the right eye9 they argued, may stimulated growth of the left hemisphere. in advance of the right and so lay the basis for asymmetry. This hypo.thesis gained support from experiments in which eggs were incubated in darkness during the last 3 days of fncubation (Rogers 1982; The: chiccks hatched from these eggs were still Zappia and Rogers TM.T)L asymmetrical in brain function, but half the population had the orientation in one direction and half in the other so that no over-all population asymmetry or bias was present. Further studies showed that as little as 2 to 3 hours of light exposure on day ?g of incubation is sufficient to orient the direction of brain asymmetry so that all individuals are asymmetrical in the same direction and a population.bias Ls present. Testosterone, alter the direction Bullock 1985). These visual discrimination copulation. 5 -dihy,drotestosterone of brain asymmetry hormones reverse the learning, and shift and oestrogen treatment can also in chickens (Zappia and Rogers 1985; direction of LES-RES asymmetry for hemispheric dominance for control of Brain asymmetry in the chicken can therefore be affected by factors under farm management control; viz9 light exposure during incubation and Various procedures used hormonal exposure during post-hatching development. in raising chickens commerically may be inadvertently manipulating asymmetry either to advantage or disadvantage for productivity or for welfare of the I believe that we need to understand these factors. Whether the chickens. chicks are hatched from egt3s incubated under light or darkness may have on behaviour in flocks, since one might expect the effects important behavioural organisation of a flock of birds with a consistent bias in the direction of individual brain asymmetries to differ from a flock in which Whether there is no bias in the direction of individual brain asymmetries. these differences would be manifest in, say, different levels of aggression, which is feasible, can now be determined. Of course, the monocular polypeeper suggested here could oniy be successfully applied, if at aI1, to chick,s with a flock bias in the direction of asymmetry. REFERENCES ANDREW, R.J. (1982). In *Advances in Vertebrate Neuroethologyy, eds J-P. Ewert and D. Ingle. (Plenum: New York) ANDREW, R.J. and BRENNAN, A. (1983). Anim. Behav. 31: 1166. ANDREW, R.J. and ROGERS, L.J. (1972). Nature, 237: 343. ANDREW, R.J., MENCH, J, and RAINEY, C. (1982). In yAnalysis of Visual eds D.J, Ingle, M.A. Goodale and Behaviour'. p. 197 Mansfield, (MIT Press: Mass,). BULLOCK, S.P. (198% Ph.D. Thesis, in preparation, Monash University. R.J.W. DENENBERG, V.H. (1981). Behav, Brain. Sci, &: 1, GASTON, LE. and GASTON, M.G. (1984). Brain Res. 303: 190. HAMBLEY, J.W. and ROGERS, L.J. (1979). Neurosci. C 677, 4: HARNARD, S.R., DOTY, D.W.9 GOLDSTEIN9 L,, JAYNES, J. and RRAUTHAMER, G., eds 'Lateralisation in the Nervous System', (Academic Press: (1977). New York). HOWARD, K.J., ROGERS, L.J. and BOURA, A,L.A. (1980). Brain Res. 188: 369, ROGERS, L.J. (1980). Bird Behav. 2: lo ROGERS, L.J. (1982). Nature (London). 297: 223. ROGERS, L.J. (198% Advances in Study of Behav. 16: S 187 0 ROGERS, L.J. and ANSON, J.M. (Ig78). Pharm. Biochem, Behav. - 735. 3: ROGERS, L.J. and ANSON, J.M, (1979). Pharm, Biochem, Behav. 10: 67% ROGERS, L.J. and HAMBLEY, J,W, (1982). Behav. Brain Res, 2: lo ROGERS, L.J., DRENNEN, HOD. and MARK9 R,F, (1974). Brain Res, B: 213. ROGERS, L.J., ZAPPIA, J,V. and BULLOCK, SOP. (1985). Experientia, in press, ZAPPIA, J.V. and ROGERS, L.J. (198% Dev, Brain Res. 11: 93. ZAPPIA, J.V. and ROGERS, L,J. (198% Submitted to Behav, Brain Res,