- Email: [email protected]
Chemosensory Learning in the Chicken Embryo
Physiology & Behavior, Vol. 64, No. 2, pp. 133–139, 1998 © 1998 Elsevier Science Inc. All rights reserved. Printed in the U.S.A. 0031-9384/98 $19.00 1 .00
PII S0031-9384(98)00037-7
Chemosensory Learning in the Chicken Embryo H. SNEDDON, R. HADDEN AND P.G. HEPPER1 School of Psychology, The Queen’s University of Belfast, Belfast BT7 1NN, UK Received 22 January 1997; Accepted 5 January 1998 SNEDDON, H., R. HADDEN AND P. G. HEPPER. Chemosensory learning in the chicken embryo. PHYSIOL BEHAV 64(2) 133–139, 1998.— Prenatal chemosensory learning has been demonstrated in mammals, fish, amphibians, and insects, but not birds, although there is evidence of the avian’s ability to learn auditory stimuli before hatching. This paper examines how exposure to a chemosensory stimulus (strawberry) prior to hatching affects subsequent chemosensory preferences of newly hatched chicks. The chicks’ preferences were assessed at 2 days after hatching using an “olfactory” preference test (strawberry-smelling shavings versus water-coated shavings) and at 4 days after hatching using a “gustatory” preference test (strawberry-flavoured water versus unflavoured water). Chicken embryos were exposed to strawberry from Day 15 to Day 20 of incubation by either presenting the odour in the air around the egg, rubbing it onto the shell, or injecting it into the air space. With no exposure to strawberry before hatching, strawberry was highly aversive to chicks after hatching. However, following exposure to strawberry before hatching, chicks expressed a greater preference for (or weaker aversion to) the strawberry stimulus. Chicks exposed to strawberry before hatching drank more strawberryflavoured water and spent more time in a strawberry-scented area than chicks having no exposure before hatching. This change in preference was specific to the stimulus experienced before hatching and was present in the absence of any posthatching exposure to the stimulus. The results demonstrate that a chick’s chemosensory preferences are changed as a result of experience with a stimulus before hatching and are suggestive of learning. The results, similar to those obtained in other animal groups, indicate the universality of “prenatal” chemosensory learning in the animal kingdom. A possible role of embryonic chemosensory learning for recognition is discussed. © 1998 Elsevier Science Inc. Embryo
Learning
Chemosensory
Chicken
Bird
Olfactory
Gustatory
(24)). An earlier study demonstrated that Day 16 chicken embryos could form a rudimentary conditioned swallowing reflex using a 3000-Hz tone as the conditioned stimulus and saccharin as the unconditioned stimulus (23). However, it is not clear whether this conditioning relied on the taste and smell of saccharin or the tactile and/or mechanical stimuli applied during the procedure. Bird embryos do respond to odours immediately prior to hatching once the beak has penetrated the air space (29,32). Furthermore, newly hatched chicks show evidence of both exposure (16) and taste aversion learning (30) to olfactory stimuli. Air readily diffuses across the shell of the egg (3) so there is a possibility that odours present in the air, or on the egg shell itself, may penetrate the shell and reach the embryo. Our study examined the ability of the chicken embryo to learn about chemosensory stimuli introduced into its environment before hatching.
THE ability of animals to learn prenatally has now been widely documented. Studies of the rat fetus, for example, have demonstrated exposure learning (9,25), habituation (28), taste-aversion learning (26), and classical conditioning (27). However, the list of animal species in which prenatal learning has been found is not restricted to mammals. Birds (24), amphibians (12), fish (6), and even insects (4,15) have all been shown to learn prenatally. The majority of studies of prenatal learning have examined exposure learning. In these studies, animals are exposed to a stimulus (usually chemosensory or auditory) before birth. After birth, their response to this familiar stimulus is compared either to their response to an unfamiliar stimulus or to the response of other individuals who were not exposed to the familiar stimulus prenatally (9,12,24,25). Chemosensory stimuli have been the stimuli of choice in studies of prenatal learning in insects (4,15), fish (6), amphibians (12), and nonhuman mammals (22). For studies of humans (11) (and also birds (14,18,24)) the stimulus of choice has been sound, although it has been demonstrated that the human fetus can learn about odours and tastes (10). These findings suggest that embryonic learning of chemosensory stimuli by simple exposure may be common to all animal groups. To date there have been no studies of embryonic chemosensory learning in birds, although it has been shown that birds have the ability to learn auditory stimuli before hatching (e.g., 1
Exposure
EXPERIMENT 1 METHOD
Subjects Fertilised eggs of domestic fowl (Ross poultry, R7) obtained from Moy Park were used. These were incubated at 37.4 °C with a relative humidity of 60%. Following hatching (21–22 days after fertilisation), all chicks were housed in their experimental groups
To whom requests for reprints should be addressed. E-mail: [email protected]
133
134 (see below) in plastic cages, 55.6 cm (l) 3 39 cm (w) 3 19 cm (h), with a wire top. The floor was covered with wood shavings. Food (Jordan’s starter feed for chicks) and water were freely available. Chicks were group housed in these cages in a separate room from the one in which they had been incubated, thus preventing exposure to the stimulus experienced before hatching. Chicks remained in this cage until testing. Treatment The fertilised eggs were divided into four groups depending upon their treatment before hatching, i.e., the method of exposure to the novel chemosensory stimulus. All treatments started on Day 15 of incubation and ended on Day 20. There was no experimental exposure to the stimulus on Day 21. The stimulus used was strawberry. This study used Strawberry Flavouring, a publicly available food additive, made by Supercook (Supercook, Sherburn-in-Elmet, Leeds, England). The solution was used undiluted. At any one time only a single group was present in the incubator, thereby ensuring eggs and birds from different exposure conditions were not mixed together. Group 1: Exposure via the Air. Thirty-four eggs were treated before hatching; 5 failed to hatch and thus the final sample was 29. Two 10-mL plastic dishes, 2 cm in diameter, were placed at opposite sides of the incubator. Five milliliters of undiluted strawberry solution was placed in each dish. The heat from the incubator circulated the odour around the incubator and the smell was readily identifiable when the top of the incubator was removed. Each day the fluid was removed and replaced with a fresh solution (5 mL per dish). The dishes and solution were removed permanently at the beginning of Day 21, at which time none of the chicks had hatched. All chicks hatched between Days 21 and 22 and were immediately transferred, within 60 min of hatching, from the incubation room into the new holding cages in a separate room. Group 2: Exposure via the Shell. Thirty-four eggs were exposed to the strawberry stimulus before hatching; 6 eggs failed to hatch and thus the final sample was 28. In this group the strawberry was applied directly onto the shell of the egg. Eggs were removed individually from the incubator, and 1 mL of the strawberry solution was gently rubbed onto the shell surface with a nonabsorbent cloth. The egg was then re-placed in the incubator. This procedure was repeated from Day 15 to Day 20. All chicks were then treated identically to those in Group 1. Group 3: Exposure via direct injection into the Egg. Of 34 eggs treated before hatching, 7 failed to hatch and thus the final sample was 27. The position of the air space was identified and a 0.75-mm hole was drilled above this through the shell with a dentist’s drill. Strawberry solution (0.05 mL) was then injected into the air space using a syringe. The hole was sealed with a small piece of cellotape to prevent the fluid from re-emerging. This procedure was repeated daily from Day 15 to Day 20. All chicks were then treated identically to those in Group 1. Group 4: No exposure. Thirty-four eggs were used, but 9 failed to hatch and the final sample size was 25. Eggs were not exposed to any odour. All chicks hatched between Days 21 and 22 and were then treated identically to chicks in Group 1.
SNEDDON, HADDEN AND HEPPER following hatching. For this test, half of the cage floor was covered with shavings scented with the strawberry solution (5 mL of the strawberry solution was shaken in a bag with the shavings before being placed on the floor of the cage), and the other half with shavings covered with water (5 mL of water was shaken in a bag with the shavings before being placed on the floor of the cage). Chicks walked on the scented and unscented shavings on the floor of the cage. Attempts to test chicks individually revealed that single chicks became extremely distressed in the test cage. This was overcome by testing two chicks from the same exposure condition together. Chicks were placed in the centre of the cage and left for 1 min to habituate to the new surroundings. The trial then commenced and lasted for 3 min. During this time the amount of time each chick spent in the strawberry half of the cage was recorded in seconds. As chicks were tested in pairs, the appropriate measure for the analysis was the mean time each pair of chicks spent in the strawberry half of the cage. This was calculated by summing the time spent on the strawberry side of the cage for both chicks and dividing by 2. Thus each pair of chicks provided a single score for the analysis. “Gustatory” preference test. Chicks were again tested in a cage identical to the one in which they were housed after hatching. In this test the cage was covered with unaltered wood shavings. Chicks were tested in groups of four. Two 250-mL drinking bottles were placed at each end of the cage. One bottle contained water which had been flavoured with strawberry (248 mL of water 1 2 mL of strawberry solution), and the other contained “pure” water. Chicks were left with food (located in the centre of the cage) freely available in this environment for 24 h. Six groups of four chicks were tested from each treatment condition. For half of the tests the strawberry bottles were on the left; for the other half of the tests they were on the right. The amount of water drunk from each bottle (flavoured and unflavoured) was recorded for each group of chickens. The amount of strawberry water drunk was expressed as a percentage of the total amount of water drunk (i.e., strawberry water/[strawberry 1 unflavoured water] 3100). Thus each group of four chicks produced a single score for the analysis, percentage strawberry-flavoured water consumed. RESULTS
“Olfactory” Preference Test The effect of exposure to strawberry before hatching on the “olfactory” preference of 2-day-old chicks was assessed by a one-way between-subjects analysis of variance for the factor of exposure (Air, Shell, Egg, None). There was a highly significant effect of exposure, F(3, 51) 5 9.007, p , 0.0001 (see Fig. 1). Posthoc Neuman–Keuls tests revealed that a significantly greater amount of time (p , 0.01) was spent in the strawberry side of the cage by chicks exposed to the odour by the air (Group 1: mean time (seconds) 6 SD 79.4 6 51.4), by their shell (Group 2: 93.4 6 45.4), or by direct injection into the egg (Group 3: 76.5 6 57.9) than by the chicks not exposed to the odour before hatching (Group 4: 10.8 6 9.4). There were no differences in the time spent in the strawberry half of the cage between the three different routes of exposure.
Procedure
“Gustatory” Preference Test
Chicks were given two preference tests after hatching. All chicks were given an “olfactory” preference test 2 days after hatching and a “gustatory” preference test 4 days after hatching. “Olfactory” preference test. Chicks were tested in a plastic cage identical to one in which they were housed as a group
The effects of exposure to strawberry before hatching on the drinking preferences of 4-day-old chicks was assessed by a oneway between-subjects analysis of variance for the factor of exposure (Air, Shell, Egg, None). There was a significant effect of exposure, F(3, 20) 5 3.478, p 5 0.0352 (see Fig. 1). Posthoc Neuman–Keuls tests revealed that a significantly greater amount of
CHICKEN EMBRYONIC LEARNING
135 amount of strawberry-flavoured water was drunk by newly hatched chicks. EXPERIMENT 2 To ensure the above results were due to exposure to strawberry before hatching rather than arising from the additional handling or some other procedural factor, a further study was undertaken. The subjects, treatment, procedure, and analysis were identical to those reported for Experiment 1 with the single exception that the strawberry solution during the treatment phase of the experiment was replaced with water. METHODS
Group 1w: Air exposure. Of 34 eggs, 4 failed to hatch and thus the final sample was 30. Group 2w: Shell exposure. Of 34 eggs, 6 failed to hatch and thus the final sample was 28. Group 3w: Egg exposure. Of 34 eggs, 6 failed to hatch and thus the final sample was 28. Group 4w: No exposure. Of 34 eggs, 8 failed to hatch and thus the final sample was 26. RESULTS
“Olfactory” Preference Test There was no difference in the preference exhibited for strawberry among the four treatment groups, F(3, 52) 5 0.054, not significant (NS). Chicks from all four groups maintained an aversion for the strawberry-scented side of the maze (Fig. 2). “Gustatory” Preference Test FIG. 1. Mean (6SD) time in seconds spent in the “strawberry” side of the cage and mean (6SD) percentage amount of strawberry-flavoured water drunk by chicken embryos exposed to strawberry via the air (Group 1, Air), via their shell (Group 2, Shell), or by injection into the air space (Group 3, Egg) or not exposed (Group 4, None).
strawberry-flavoured water (p , 0.05) was drunk by chicks exposed to the odour via their shell (Group 2: mean percentage water drunk 6 SD 40.1 6 30.7) than chicks receiving no prehatch exposure (Group 4: 3.2 6 2.0). There were no other differences.
There was no difference between the amount of strawberry water drunk by the four groups, F(3, 20) 5 0.398, NS. Chicks from all four groups maintained their aversion to strawberry-flavoured water (Fig. 2). SUMMARY
In both tests individuals maintained their natural aversion to strawberry. This indicates that the manipulations performed on the eggs to expose the embryos to strawberry were not responsible for the increased preference for strawberry shown in Experiment 1. Rather this change in preference resulted directly from experience of the strawberry solution.
SUMMARY
EXPERIMENT 3
Exposure to strawberry before hatching changed the response of chicks to this stimulus after hatching. It is worth noting the valence of the chicks’ response. Chicks not previously exposed to the stimulus regarded this as aversive. Exposure to the stimulus before hatching changed this response. Chicks now spent more time on the side of the cage with the strawberry odour and drank more water flavoured with the strawberry. Exactly how exposure before hatching influences this change in behaviour is unknown. It may reduce the aversiveness of odours or increase attraction to a familiar odour. Both processes would result in increased time spent in proximity to strawberry odour or drinking strawberry-flavoured water, but the underlying mediation of this remains elusive. With no exposure to strawberry before hatching, the odour of strawberry or water flavoured with strawberry was aversive to chicks after birth. However, following exposure to strawberry before hatching a significantly greater time was spent on the strawberry-scented side of the cage and a significantly greater
Exposure to strawberry before hatching changed the chick’s response to strawberry after hatching. One factor which needs to be considered is the specificity of this change in preference. Does exposure to a chemosensory stimulus before hatching result in a change in preference to all novel odours and tastes or is the change specific to the stimulus experienced before hatching? It may be that prehatch exposure to a chemosensory stimulus results in a nonspecific change in the chick’s preference for all chemosensory stimuli rather than specifically for the odour experienced before hatching. To determine the specificity of the preference found in Experiment 1, chicks were exposed before hatching to strawberry and examined after hatching in one of two choice tests: chicks were given either a choice test between vanilla and water or a choice test between strawberry and vanilla. As the shell exposure condition resulted in the strongest preference, only this treatment was studied below. If exposure to the stimulus before hatching changed the chick’s preference to all chemosensory stimuli, it
136
SNEDDON, HADDEN AND HEPPER the strawberry solution was replaced by vanilla (again this was a publicly available food additive, Vanilla Flavouring, manufactured by Supercook, Sherburn-in-Elmet, Leeds, UK). The vanilla was diluted with distilled water such that it was perceived by the experimenters to be of the same intensity as the strawberry solution used in Experiment 1. Second, 20 chicks who were exposed to strawberry before hatching and 22 who were not were examined in the “olfactory” and “gustatory” tests and given the choice between strawberry and vanilla. These tests were performed identically to those reported in Experiment 1 with the single exception that the water was replaced by vanilla, giving the chicks a choice between strawberry and vanilla. RESULTS
Vanilla versus Water “Olfactory” preference test. There was no significant difference [t(24) 5 1.171, NS] between the time (seconds) spent on the vanilla side of the cage by chicks exposed to strawberry before hatching (mean 6 SD 15.7 6 28.5) and those not so exposed (5.9 6 10). “Gustatory” preference test. There was no significant difference [t(10) 5 0.001, NS] between the percentage amount of vanilla-flavoured water and unflavoured water drunk by chicks exposed to strawberry before hatching (mean 6 SD 3.2 6 2.3) and those not so exposed (3.2 6 2.1). Strawberry versus Vanilla
FIG. 2. Mean (6SD) time in seconds spent in the “strawberry” side of the cage and mean (6SD) percentage amount of strawberry-flavoured water drunk by chicken embryos exposed to water via the air (Group 1w, Air), via their shell (Group 2w, Shell), or by injection into the air space (Group 3w, Egg) or not exposed (Group 4w, None).
would be expected that chicks exposed to strawberry before hatching would show a different response to vanilla in the vanilla versus water test compared to chicks given no exposure to strawberry before hatching. Furthermore, if the change in preference was nonspecific to the stimulus experienced before birth, i.e., the strawberry, chicks may show no preference for strawberry when given the choice between strawberry and vanilla. METHOD
Subjects Treatment conditions were the same as reported above for Groups 2 and 4 of Experiment 1. From an initial 52 eggs in each group, 46 hatched in the strawberry-exposed group and 48 in the control, no strawberry exposure, group and formed the subjects for this experiment.
“Olfactory” preference test. There was a significant difference [t(19) 5 3.088, p 5 0.0061] between the time (seconds) spent on the strawberry side of the cage by chicks exposed to strawberry before hatching (mean 6 SD 125.5 6 37.5) and those not so exposed (76.8 6 34.7). Paired t-tests to examine the time spent on the strawberry and vanilla sides of the cage for each group were then undertaken. Chicks exposed to strawberry before hatching spent a significantly [t(9) 5 2.797, p 5 0.0208] greater amount of time on the strawberry side (mean 6 SD 125.5 6 37.5) of the cage than on the vanilla side (54.5 6 37.5). Chicks not exposed to strawberry did not differ [t(10) 5 1.26, NS] in the time they spent on the strawberry (mean 6 SD 76.8 6 34.7) or vanilla side of the cage (103.2 6 34.7). “Gustatory” preference test. There was a significant difference [t(8) 5 2.875, p 5 0.0207] between the percentage amount of strawberry-flavoured and vanilla-flavoured water drunk by the chicks. Chicks exposed to strawberry before hatching drank more strawberry-flavoured water (mean 6 SD 78 6 13.8) than those not exposed to strawberry before birth (45.7 6 20). Paired t-tests to examine the percentage amount of strawberry-flavoured and vanilla-flavoured water drunk by each group were then undertaken. Chicks exposed to strawberry before hatching drank significantly [t(4) 5 4.523, p 5 0.0106] more strawberryflavoured water (mean 6 SD 78 6 13.8) than vanilla-flavoured water (22 6 13.8). Chicks not exposed to strawberry did not differ [t(4) 5 0.46, NS] in the amount of strawberry-flavoured water (mean 6 SD 45.7 6 20) or vanilla-flavoured water they drank (54.3 6 20).
Procedure Two choice tests were undertaken in this study. First, 26 chicks from each group were examined in the “olfactory” and “gustatory” tests and given the choice between vanilla or water. Both the “olfactory” preference test (Day 2 posthatch) and the “gustatory” test (Day 4 posthatch) were performed identically to the tests described above in Experiment 1. The exception to this was that
SUMMARY
Exposure to strawberry before hatching results in a preference that did not affect the chick’s response to a novel, vanilla in this case, chemosensory stimulus. Further when given a preference test between strawberry and vanilla, only those chicks exposed to strawberry before hatching showed a preference for strawberry.
CHICKEN EMBRYONIC LEARNING These results indicate that the preference acquired by the chicks to the strawberry in Experiment 1 would appear to be specific to stimulus experienced before hatching. EXPERIMENT 4 One problem with the procedure adopted to rear the chicks in Experiment 1 is that the chicks may experience the stimulus between the time of hatching and the time of testing. Exposure to the stimulus stopped on Day 20 to prevent this possibility. However, although the strawberry solution was not applied on Day 21, traces may still have remained from previous treatments, especially in the groups exposed to the stimulus by rubbing it on their shells or by injection into the egg. It is possible in these groups that the chicks could have acquired the preference for strawberry in the period between hatching and removal from the incubator from smelling the strawberry on the shell or by eating the shell either during or after hatching. To rule out this possibility, a further study was undertaken in which eggs at Day 21 were artificially hatched and removed to a new cage and room, preventing exposure to the stimulus after hatching. Subjects Treatment conditions were the same as reported above for Groups 2 (n 5 22) and 4 (n 5 24) of Experiment 1. The only exception to the treatment undertaken by these eggs was that on Day 21 eggs were closely watched. When cracks in the shell were noted, but prior to the appearance of a hole, the egg was removed from the incubator, the shell was opened, and the chick was hatched and placed in a new cage. This ensured the chick had no experience of the strawberry stimulus once hatched prior to testing. Procedure Both the “olfactory” preference test (Day 2 posthatch) and the “gustatory” preference test (Day 4 posthatch) were performed identically as described in Experiment 1. RESULTS
“Olfactory” Preference Test There was a significant difference [t(22) 5 5.01, p 5 0.001] between the time (seconds) spent on the strawberry side of the maze by pairs of chicks exposed before hatching to strawberry (mean 6 SD 71.3 6 38.5) and those not so exposed (10.8 6 13.2). “Gustatory” Preference Test For this test there were five groups of four chicks from the strawberry exposure group and six from the control, no exposure, condition. The results indicated there was a significant difference [t(9) 5 3.209, p 5 0.011] between the percentage amount of strawberry water drunk by chicks exposed before hatching to strawberry (mean 6 SD 21.6 6 13.6) and those not so exposed (3.7 6 2.3). SUMMARY
In both “olfactory” and “gustatory” tasks chicks exposed to strawberry before hatching showed a greater preference for this stimulus than chicks not so exposed. Thus exposure to a chemosensory stimulus before hatching alone is sufficient to change the preferences of chicks after hatching.
137 DISCUSSION Previous studies have indicated the chicken is able to learn auditory stimuli before hatching. The results of this study extend these findings by demonstrating that the chicken is able to learn chemosensory stimuli before hatching. Experimental exposure to the stimuli in this study took place over 5 days before hatching. However, it is possible that in the shell and injection groups the embryo could have experienced the stimulus on Day 21 from its continued presence on the shell or inside the egg. Thus it is reasonable to assume that at some point during this time (E15–E21, E 5 embryonic day) the chicken embryo’s olfactory and/or gustatory senses commence functioning. Previous studies have demonstrated a functioning chemosensory sense at E21 (29). Our results from the air exposure group, where exposure was confined to E15–E20, suggest this may be functioning even earlier. The olfactory system begins its development early in chick embryogenesis (1,20). Outgrowths of olfactory neurons are seen at E3 and olfactory bulb morphogenesis begins at E7. Around this time proteins involved in the reception and transduction of olfactory signals are found. The olfactory epithelium is mature by E19. It is most likely that the peripheral olfactory system is mature prior to central processing of olfactory signals although exactly when the olfactory system becomes functional is unknown. The functional ontogenesis of the gustatory system is less well understood. Taste buds develop rapidly from E16 (no buds) to hatching (7) where numbers approximate half that of the adult bird. Again when they become functional is unknown. The behavioural evidence presented in this study suggests both olfactory and gustatory receptors are functional before hatching. Studies of the embryonic learning of chemosensory stimuli find it difficult to separate the roles of gustation and olfaction in exposure and the resulting preference. Two preference tests were therefore conducted in an attempt to separate the roles of smell and taste in the exhibition of the preference after hatching. Chicks exposed to strawberry before hatching exhibited a change in preference to this when tested in the “olfactory” task and in the “gustatory” task. There was undoubtedly some olfactory stimulus present in the “gustatory” test which could have guided the chick to the bottle. However, a gustatory component was most certainly involved once the chick began to drink and gustation was probably involved in the continuation of drinking following the initial consumption of the fluid. Thus the “gustatory” test may have contained some olfactory stimuli as well as gustatory stimuli. Taste stimuli played very little role in the “olfactory” test. Although chicks walked on the shavings, there was little observed pecking of the shavings by the chicks and hence little opportunity for gustatory stimuli in this preference. However, the exact role of odour and/or taste in the preferences observed has yet to be determined. At the exposure stage both taste and smell may have been involved. The chick begins to swallow amniotic fluid from about halfway through incubation (21) and possibly earlier (17). The nares are blocked by tissue for most of the incubation period; this degenerates toward the end of this period and the nares are open by E20 (29). Whilst this may restrict the access of stimuli directly to the olfactory receptors, the swallowing activities of the embryo may stimulate the olfactory receptors by moving fluid from the mouth to the nasal cavities. Such fluid movement is observed in the human fetus from midway through gestation (P.G.H., personal observation) and may occur in the chicken embryo. The gustatory sense however may have greater direct access to chemosensory stimuli. It should be noted that
138
SNEDDON, HADDEN AND HEPPER
exposure before hatching resulted in both an “olfactory” and a “gustatory” preference. If no olfactory exposure occurred before hatching, this would indicate intermodal transfer of information between gustatory and olfactory senses. Olfactory processes in birds have often been viewed as less important than other sensory processes, especially auditory, and have by comparison been little studied. However, recent years have seen renewed interest in avian olfaction. Examination of olfactory thresholds reveals birds (passerines) have detection thresholds similar to those of rats (5). Olfaction has been postulated to play a role in homing (2,33). The role of olfaction in feeding behaviour has aroused considerable interest (19,30). The studies reported here indicate that birds have a functioning chemosensory system as embryos and suggest a role for chemosensation in the perinatal period. The methods of exposure used in the study aimed to mimic any “natural” way that the chicken embryo may have contact with chemosensory stimuli. Whilst the direct injection of stimuli into the egg was purely experimental, exposure to stimuli from the air surrounding the egg or from its presence on the egg shell may both occur naturally during the embryo’s time in the egg. What function the learning of chemosensory stimuli before hatching may have for the chick is unknown. In other species prenatal chemosensory exposure may influence subsequent recognition and/or food preferences (9) and it is possible it may serve similar functions in birds. Visual imprinting and auditory imprinting have been well established in newly hatched chicks (13), and there is some evidence for olfactory imprinting in chicks (31). In the embryo there is much evidence for auditory imprinting (8,14,18). Rogers has speculated that the chicken embryo may imprint on the odour of its surroundings, e.g., from the nest or hen (20). It is possible that odours from the nest or the mother when incubating her eggs would pass across the shell and stimulate the chemosensory receptors of the chicken embryo.
Indeed such stimuli would have access to the embryo from well before hatching and this may increase the concentration of the chemosensory stimulus in the egg. Hence, if the olfactory or gustatory system is functioning before hatching, the possibility of embryonic olfactory imprinting for maternal or nest recognition should be considered. Six-day-old chicks show a preference for their own soiled nest litter over strange or clean bedding (16). However, given the 6 days of exposure after hatching, this preference cannot be attributed to embryonic exposure. This study adds to the growing list of species in which prenatal chemosensory learning has been established. Olfaction is regarded, in evolutionary terms, as the oldest sense and its functioning across different species in the perinatal period shows remarkable similarity and universality. The finding of “prenatal” chemosensory learning across major animal groups may suggest some evolutionary importance to this ability perhaps ensuring recognition of parent(s), kin, nestmates, or the “nest site” or influencing feeding behaviour (11). Exposure of chicken embryos to a chemosensory stimulus affects their subsequent preference for this stimulus after birth. The change in preference is specific to the stimuli experienced before hatching. The chicken embryo may experience chemosensory stimuli naturally during the course of incubation, and chemosensory learning may have some function, possibly enabling recognition in the newly hatched chick. Studies are currently underway to examine the function of chemosensory stimuli in the perinatal period in the chick. ACKNOWLEDGEMENTS
We acknowledge the support of Prof. K Brown, School of Psychology, for providing facilities and Ken McCarroll for husbandry support. All studies were performed under license issued by Department of Health and Social Services (DHSS). The comments and suggestions of the reviewers have greatly improved this paper.
REFERENCES 1. Ayer-Le Lie`vre, C.; Lapointe, F.; Leibovici, M. Avian olfactory neurogenesis. Biol. Cell. 84:25–34; 1995. 2. Bingman, V. P.; Benvenuti, S. Olfaction and homing ability of pigeons in the Southeastern United States. J. Exp. Zool. 276:186 –192; 1996. 3. Board, R. G. Properties of avian egg shells and their adaptive value. Biol. Rev. 57:1–28; 1982. 4. Caubet, Y.; Jaisson, P.; Lenoir, A. Preimaginal induction of adult behaviour in insects. Q. J. Exp. Psychol. 44B:165–178; 1992. 5. Clark, L.; Avilova, K. V.; Bean, N. J. Odor thresholds in passerines. Comp. Biochem. Pysiol. A Physiol. 104:305–312; 1993. 6. Courtenay, S. C. Learning and memory of chemosensory stimuli by underyearling coho salmon Oncorhynchus kisutch (Walbaum). Vancouver, BC: University of British Columbia, 1989 (PhD Thesis). 7. Ganchrow, J. R.; Ganchrow, D. Development of taste buds in the chick-embryo (Gallus gallus domesticus). Ann. NY Acad. Sci. 510: 307–308; 1987. 8. Gottlieb, G. Development of species identification in ducklings: IX. The necessity of experiencing normal variations in embryonic auditory stimulation. Dev. Psychobiol. 15:507–517; 1982. 9. Hepper, P. G. Adaptive fetal learning: Prenatal exposure to garlic affects postnatal preferences. Anim. Behav. 36:935–936; 1988. 10. Hepper, P. G. Human fetal “olfactory” learning. Int. J. Prenatal Perinatal Psycho. Med. 7:147–151; 1995. 11. Hepper, P. G. Fetal memory: Does it exist? What does it do? Acta Paediatr. Suppl. 416:16 –20; 1996. 12. Hepper, P. G.; Waldman, B. Embryonic olfactory learning in frogs. Q. J. Exp. Psychol. 44B:179 –197; 1992. 13. Horn, G. Imprinting, memory and the brain. Oxford: Clarendon Press; 1985.
14. Impekoven, M. Responses of laughing gull chicks (Larus atricilla) to parental attraction and alarm calls, and effects of prenatal auditory experience on the responsiveness to such calls. Behaviour. 61:250 – 278; 1976. 15. Isingrini, M.; Lenoir, A.; Jaisson, P. Pre-imaginal learning as a basis of colony-brood recognition in the ant, Cataglyphis cursor. Proc. Natl. Acad. Sci. USA. 82:8545– 8547; 1985. 16. Jones, R. B.; Gentle, M. J. Olfaction and behavioural modifications in domestic chicks (Gallus domesticus). Physiol. Behav. 43:917–924; 1985. 17. Kuo, Z. Y. Ontogeny of embryonic behaviour in Aves. I. The chronology and general nature of behaviour in the chick embryo. J. Exp. Zool. 61:395– 430; 1932. 18. Lickliter, R.; Stoumbos, J. Modification of prenatal auditory experience alters postnatal auditory preferences of bobwhite quail chicks. Q. J. Exp. Psychol. 44B:199 –214; 1992. 19. Marples, N. M.; Roper, T. J. Effects of novel colour and smell on the response of naive chicks towards food and water. Anim. Behav. 51:1417–1424; 1996. 20. Rogers, L. J. The development of brain and behaviour in the chicken. Cambridge: CAB International; 1995. 21. Romanoff, A. L. The avian embryo. New York: Macmillan; 1960. 22. Schaal, B.; Orgeur, P. Olfaction in utero: Can the rodent model be generalized? Q. J. Exp. Psychol 44B:245–278; 1992. 23. Sedlacek, J. Functional characteristics of the centre of the unconditioned reflex in elaboration of a temporary connection in chick embryos. Physiol. Bohemoslov. 11:313–318; 1962. 24. Shindler, K. M. A three year study of fetal auditory imprinting. J. Wash. Acad. Sci. 74:121–124; 1984.
CHICKEN EMBRYONIC LEARNING 25. Smotherman, W. P. In utero chemosensory experience alters taste preferences and corticosterone responsiveness. Behav. Neural Biol. 36:61– 68; 1982. 26. Smotherman, W. P. Odor aversion learning by the rat fetus. Physiol. Behav. 29:769 –771; 1982. 27. Smotherman, W. P.; Robinson, S. R. The rat fetus in its environment. Behavioral adjustments to novel, familiar, aversive and conditioned stimuli presented in utero. Behav. Neurosci. 99:521–530; 1985. 28. Smotherman, W. P.; Robinson, S. R. Habituation in the rat fetus. Q. J. Exp. Psychol. 44B:215–230; 1992.
139 29. Tolhurst, B. E.; Vince, M. A. Sensitivity to odours in the embryo of the domestic fowl. Anim. Behav. 24:772–779; 1976. 30. Turro, I.; Porter, R. H.; Picard, M. Olfactory cues mediate food selection by young chicks. Physiol. Behav. 55:761–767; 1994. 31. Vallortigara, G.; Andrew, R. J. Olfactory lateralization in the chick. Neuropsychologia. 32:417– 423; 1994. 32. Vince, M. A. Taste sensitivity in the embryo of the domestic fowl. Anim. Behav. 25:797– 805; 1977. 33. Wiltschko, R. The function of olfactory input in pigeon orientation. Does it provide navigational information or play another role? J. Exp. Biol. 199:113–119; 1996.
PII S0031-9384(98)00037-7
Chemosensory Learning in the Chicken Embryo H. SNEDDON, R. HADDEN AND P.G. HEPPER1 School of Psychology, The Queen’s University of Belfast, Belfast BT7 1NN, UK Received 22 January 1997; Accepted 5 January 1998 SNEDDON, H., R. HADDEN AND P. G. HEPPER. Chemosensory learning in the chicken embryo. PHYSIOL BEHAV 64(2) 133–139, 1998.— Prenatal chemosensory learning has been demonstrated in mammals, fish, amphibians, and insects, but not birds, although there is evidence of the avian’s ability to learn auditory stimuli before hatching. This paper examines how exposure to a chemosensory stimulus (strawberry) prior to hatching affects subsequent chemosensory preferences of newly hatched chicks. The chicks’ preferences were assessed at 2 days after hatching using an “olfactory” preference test (strawberry-smelling shavings versus water-coated shavings) and at 4 days after hatching using a “gustatory” preference test (strawberry-flavoured water versus unflavoured water). Chicken embryos were exposed to strawberry from Day 15 to Day 20 of incubation by either presenting the odour in the air around the egg, rubbing it onto the shell, or injecting it into the air space. With no exposure to strawberry before hatching, strawberry was highly aversive to chicks after hatching. However, following exposure to strawberry before hatching, chicks expressed a greater preference for (or weaker aversion to) the strawberry stimulus. Chicks exposed to strawberry before hatching drank more strawberryflavoured water and spent more time in a strawberry-scented area than chicks having no exposure before hatching. This change in preference was specific to the stimulus experienced before hatching and was present in the absence of any posthatching exposure to the stimulus. The results demonstrate that a chick’s chemosensory preferences are changed as a result of experience with a stimulus before hatching and are suggestive of learning. The results, similar to those obtained in other animal groups, indicate the universality of “prenatal” chemosensory learning in the animal kingdom. A possible role of embryonic chemosensory learning for recognition is discussed. © 1998 Elsevier Science Inc. Embryo
Learning
Chemosensory
Chicken
Bird
Olfactory
Gustatory
(24)). An earlier study demonstrated that Day 16 chicken embryos could form a rudimentary conditioned swallowing reflex using a 3000-Hz tone as the conditioned stimulus and saccharin as the unconditioned stimulus (23). However, it is not clear whether this conditioning relied on the taste and smell of saccharin or the tactile and/or mechanical stimuli applied during the procedure. Bird embryos do respond to odours immediately prior to hatching once the beak has penetrated the air space (29,32). Furthermore, newly hatched chicks show evidence of both exposure (16) and taste aversion learning (30) to olfactory stimuli. Air readily diffuses across the shell of the egg (3) so there is a possibility that odours present in the air, or on the egg shell itself, may penetrate the shell and reach the embryo. Our study examined the ability of the chicken embryo to learn about chemosensory stimuli introduced into its environment before hatching.
THE ability of animals to learn prenatally has now been widely documented. Studies of the rat fetus, for example, have demonstrated exposure learning (9,25), habituation (28), taste-aversion learning (26), and classical conditioning (27). However, the list of animal species in which prenatal learning has been found is not restricted to mammals. Birds (24), amphibians (12), fish (6), and even insects (4,15) have all been shown to learn prenatally. The majority of studies of prenatal learning have examined exposure learning. In these studies, animals are exposed to a stimulus (usually chemosensory or auditory) before birth. After birth, their response to this familiar stimulus is compared either to their response to an unfamiliar stimulus or to the response of other individuals who were not exposed to the familiar stimulus prenatally (9,12,24,25). Chemosensory stimuli have been the stimuli of choice in studies of prenatal learning in insects (4,15), fish (6), amphibians (12), and nonhuman mammals (22). For studies of humans (11) (and also birds (14,18,24)) the stimulus of choice has been sound, although it has been demonstrated that the human fetus can learn about odours and tastes (10). These findings suggest that embryonic learning of chemosensory stimuli by simple exposure may be common to all animal groups. To date there have been no studies of embryonic chemosensory learning in birds, although it has been shown that birds have the ability to learn auditory stimuli before hatching (e.g., 1
Exposure
EXPERIMENT 1 METHOD
Subjects Fertilised eggs of domestic fowl (Ross poultry, R7) obtained from Moy Park were used. These were incubated at 37.4 °C with a relative humidity of 60%. Following hatching (21–22 days after fertilisation), all chicks were housed in their experimental groups
To whom requests for reprints should be addressed. E-mail: [email protected]
133
134 (see below) in plastic cages, 55.6 cm (l) 3 39 cm (w) 3 19 cm (h), with a wire top. The floor was covered with wood shavings. Food (Jordan’s starter feed for chicks) and water were freely available. Chicks were group housed in these cages in a separate room from the one in which they had been incubated, thus preventing exposure to the stimulus experienced before hatching. Chicks remained in this cage until testing. Treatment The fertilised eggs were divided into four groups depending upon their treatment before hatching, i.e., the method of exposure to the novel chemosensory stimulus. All treatments started on Day 15 of incubation and ended on Day 20. There was no experimental exposure to the stimulus on Day 21. The stimulus used was strawberry. This study used Strawberry Flavouring, a publicly available food additive, made by Supercook (Supercook, Sherburn-in-Elmet, Leeds, England). The solution was used undiluted. At any one time only a single group was present in the incubator, thereby ensuring eggs and birds from different exposure conditions were not mixed together. Group 1: Exposure via the Air. Thirty-four eggs were treated before hatching; 5 failed to hatch and thus the final sample was 29. Two 10-mL plastic dishes, 2 cm in diameter, were placed at opposite sides of the incubator. Five milliliters of undiluted strawberry solution was placed in each dish. The heat from the incubator circulated the odour around the incubator and the smell was readily identifiable when the top of the incubator was removed. Each day the fluid was removed and replaced with a fresh solution (5 mL per dish). The dishes and solution were removed permanently at the beginning of Day 21, at which time none of the chicks had hatched. All chicks hatched between Days 21 and 22 and were immediately transferred, within 60 min of hatching, from the incubation room into the new holding cages in a separate room. Group 2: Exposure via the Shell. Thirty-four eggs were exposed to the strawberry stimulus before hatching; 6 eggs failed to hatch and thus the final sample was 28. In this group the strawberry was applied directly onto the shell of the egg. Eggs were removed individually from the incubator, and 1 mL of the strawberry solution was gently rubbed onto the shell surface with a nonabsorbent cloth. The egg was then re-placed in the incubator. This procedure was repeated from Day 15 to Day 20. All chicks were then treated identically to those in Group 1. Group 3: Exposure via direct injection into the Egg. Of 34 eggs treated before hatching, 7 failed to hatch and thus the final sample was 27. The position of the air space was identified and a 0.75-mm hole was drilled above this through the shell with a dentist’s drill. Strawberry solution (0.05 mL) was then injected into the air space using a syringe. The hole was sealed with a small piece of cellotape to prevent the fluid from re-emerging. This procedure was repeated daily from Day 15 to Day 20. All chicks were then treated identically to those in Group 1. Group 4: No exposure. Thirty-four eggs were used, but 9 failed to hatch and the final sample size was 25. Eggs were not exposed to any odour. All chicks hatched between Days 21 and 22 and were then treated identically to chicks in Group 1.
SNEDDON, HADDEN AND HEPPER following hatching. For this test, half of the cage floor was covered with shavings scented with the strawberry solution (5 mL of the strawberry solution was shaken in a bag with the shavings before being placed on the floor of the cage), and the other half with shavings covered with water (5 mL of water was shaken in a bag with the shavings before being placed on the floor of the cage). Chicks walked on the scented and unscented shavings on the floor of the cage. Attempts to test chicks individually revealed that single chicks became extremely distressed in the test cage. This was overcome by testing two chicks from the same exposure condition together. Chicks were placed in the centre of the cage and left for 1 min to habituate to the new surroundings. The trial then commenced and lasted for 3 min. During this time the amount of time each chick spent in the strawberry half of the cage was recorded in seconds. As chicks were tested in pairs, the appropriate measure for the analysis was the mean time each pair of chicks spent in the strawberry half of the cage. This was calculated by summing the time spent on the strawberry side of the cage for both chicks and dividing by 2. Thus each pair of chicks provided a single score for the analysis. “Gustatory” preference test. Chicks were again tested in a cage identical to the one in which they were housed after hatching. In this test the cage was covered with unaltered wood shavings. Chicks were tested in groups of four. Two 250-mL drinking bottles were placed at each end of the cage. One bottle contained water which had been flavoured with strawberry (248 mL of water 1 2 mL of strawberry solution), and the other contained “pure” water. Chicks were left with food (located in the centre of the cage) freely available in this environment for 24 h. Six groups of four chicks were tested from each treatment condition. For half of the tests the strawberry bottles were on the left; for the other half of the tests they were on the right. The amount of water drunk from each bottle (flavoured and unflavoured) was recorded for each group of chickens. The amount of strawberry water drunk was expressed as a percentage of the total amount of water drunk (i.e., strawberry water/[strawberry 1 unflavoured water] 3100). Thus each group of four chicks produced a single score for the analysis, percentage strawberry-flavoured water consumed. RESULTS
“Olfactory” Preference Test The effect of exposure to strawberry before hatching on the “olfactory” preference of 2-day-old chicks was assessed by a one-way between-subjects analysis of variance for the factor of exposure (Air, Shell, Egg, None). There was a highly significant effect of exposure, F(3, 51) 5 9.007, p , 0.0001 (see Fig. 1). Posthoc Neuman–Keuls tests revealed that a significantly greater amount of time (p , 0.01) was spent in the strawberry side of the cage by chicks exposed to the odour by the air (Group 1: mean time (seconds) 6 SD 79.4 6 51.4), by their shell (Group 2: 93.4 6 45.4), or by direct injection into the egg (Group 3: 76.5 6 57.9) than by the chicks not exposed to the odour before hatching (Group 4: 10.8 6 9.4). There were no differences in the time spent in the strawberry half of the cage between the three different routes of exposure.
Procedure
“Gustatory” Preference Test
Chicks were given two preference tests after hatching. All chicks were given an “olfactory” preference test 2 days after hatching and a “gustatory” preference test 4 days after hatching. “Olfactory” preference test. Chicks were tested in a plastic cage identical to one in which they were housed as a group
The effects of exposure to strawberry before hatching on the drinking preferences of 4-day-old chicks was assessed by a oneway between-subjects analysis of variance for the factor of exposure (Air, Shell, Egg, None). There was a significant effect of exposure, F(3, 20) 5 3.478, p 5 0.0352 (see Fig. 1). Posthoc Neuman–Keuls tests revealed that a significantly greater amount of
CHICKEN EMBRYONIC LEARNING
135 amount of strawberry-flavoured water was drunk by newly hatched chicks. EXPERIMENT 2 To ensure the above results were due to exposure to strawberry before hatching rather than arising from the additional handling or some other procedural factor, a further study was undertaken. The subjects, treatment, procedure, and analysis were identical to those reported for Experiment 1 with the single exception that the strawberry solution during the treatment phase of the experiment was replaced with water. METHODS
Group 1w: Air exposure. Of 34 eggs, 4 failed to hatch and thus the final sample was 30. Group 2w: Shell exposure. Of 34 eggs, 6 failed to hatch and thus the final sample was 28. Group 3w: Egg exposure. Of 34 eggs, 6 failed to hatch and thus the final sample was 28. Group 4w: No exposure. Of 34 eggs, 8 failed to hatch and thus the final sample was 26. RESULTS
“Olfactory” Preference Test There was no difference in the preference exhibited for strawberry among the four treatment groups, F(3, 52) 5 0.054, not significant (NS). Chicks from all four groups maintained an aversion for the strawberry-scented side of the maze (Fig. 2). “Gustatory” Preference Test FIG. 1. Mean (6SD) time in seconds spent in the “strawberry” side of the cage and mean (6SD) percentage amount of strawberry-flavoured water drunk by chicken embryos exposed to strawberry via the air (Group 1, Air), via their shell (Group 2, Shell), or by injection into the air space (Group 3, Egg) or not exposed (Group 4, None).
strawberry-flavoured water (p , 0.05) was drunk by chicks exposed to the odour via their shell (Group 2: mean percentage water drunk 6 SD 40.1 6 30.7) than chicks receiving no prehatch exposure (Group 4: 3.2 6 2.0). There were no other differences.
There was no difference between the amount of strawberry water drunk by the four groups, F(3, 20) 5 0.398, NS. Chicks from all four groups maintained their aversion to strawberry-flavoured water (Fig. 2). SUMMARY
In both tests individuals maintained their natural aversion to strawberry. This indicates that the manipulations performed on the eggs to expose the embryos to strawberry were not responsible for the increased preference for strawberry shown in Experiment 1. Rather this change in preference resulted directly from experience of the strawberry solution.
SUMMARY
EXPERIMENT 3
Exposure to strawberry before hatching changed the response of chicks to this stimulus after hatching. It is worth noting the valence of the chicks’ response. Chicks not previously exposed to the stimulus regarded this as aversive. Exposure to the stimulus before hatching changed this response. Chicks now spent more time on the side of the cage with the strawberry odour and drank more water flavoured with the strawberry. Exactly how exposure before hatching influences this change in behaviour is unknown. It may reduce the aversiveness of odours or increase attraction to a familiar odour. Both processes would result in increased time spent in proximity to strawberry odour or drinking strawberry-flavoured water, but the underlying mediation of this remains elusive. With no exposure to strawberry before hatching, the odour of strawberry or water flavoured with strawberry was aversive to chicks after birth. However, following exposure to strawberry before hatching a significantly greater time was spent on the strawberry-scented side of the cage and a significantly greater
Exposure to strawberry before hatching changed the chick’s response to strawberry after hatching. One factor which needs to be considered is the specificity of this change in preference. Does exposure to a chemosensory stimulus before hatching result in a change in preference to all novel odours and tastes or is the change specific to the stimulus experienced before hatching? It may be that prehatch exposure to a chemosensory stimulus results in a nonspecific change in the chick’s preference for all chemosensory stimuli rather than specifically for the odour experienced before hatching. To determine the specificity of the preference found in Experiment 1, chicks were exposed before hatching to strawberry and examined after hatching in one of two choice tests: chicks were given either a choice test between vanilla and water or a choice test between strawberry and vanilla. As the shell exposure condition resulted in the strongest preference, only this treatment was studied below. If exposure to the stimulus before hatching changed the chick’s preference to all chemosensory stimuli, it
136
SNEDDON, HADDEN AND HEPPER the strawberry solution was replaced by vanilla (again this was a publicly available food additive, Vanilla Flavouring, manufactured by Supercook, Sherburn-in-Elmet, Leeds, UK). The vanilla was diluted with distilled water such that it was perceived by the experimenters to be of the same intensity as the strawberry solution used in Experiment 1. Second, 20 chicks who were exposed to strawberry before hatching and 22 who were not were examined in the “olfactory” and “gustatory” tests and given the choice between strawberry and vanilla. These tests were performed identically to those reported in Experiment 1 with the single exception that the water was replaced by vanilla, giving the chicks a choice between strawberry and vanilla. RESULTS
Vanilla versus Water “Olfactory” preference test. There was no significant difference [t(24) 5 1.171, NS] between the time (seconds) spent on the vanilla side of the cage by chicks exposed to strawberry before hatching (mean 6 SD 15.7 6 28.5) and those not so exposed (5.9 6 10). “Gustatory” preference test. There was no significant difference [t(10) 5 0.001, NS] between the percentage amount of vanilla-flavoured water and unflavoured water drunk by chicks exposed to strawberry before hatching (mean 6 SD 3.2 6 2.3) and those not so exposed (3.2 6 2.1). Strawberry versus Vanilla
FIG. 2. Mean (6SD) time in seconds spent in the “strawberry” side of the cage and mean (6SD) percentage amount of strawberry-flavoured water drunk by chicken embryos exposed to water via the air (Group 1w, Air), via their shell (Group 2w, Shell), or by injection into the air space (Group 3w, Egg) or not exposed (Group 4w, None).
would be expected that chicks exposed to strawberry before hatching would show a different response to vanilla in the vanilla versus water test compared to chicks given no exposure to strawberry before hatching. Furthermore, if the change in preference was nonspecific to the stimulus experienced before birth, i.e., the strawberry, chicks may show no preference for strawberry when given the choice between strawberry and vanilla. METHOD
Subjects Treatment conditions were the same as reported above for Groups 2 and 4 of Experiment 1. From an initial 52 eggs in each group, 46 hatched in the strawberry-exposed group and 48 in the control, no strawberry exposure, group and formed the subjects for this experiment.
“Olfactory” preference test. There was a significant difference [t(19) 5 3.088, p 5 0.0061] between the time (seconds) spent on the strawberry side of the cage by chicks exposed to strawberry before hatching (mean 6 SD 125.5 6 37.5) and those not so exposed (76.8 6 34.7). Paired t-tests to examine the time spent on the strawberry and vanilla sides of the cage for each group were then undertaken. Chicks exposed to strawberry before hatching spent a significantly [t(9) 5 2.797, p 5 0.0208] greater amount of time on the strawberry side (mean 6 SD 125.5 6 37.5) of the cage than on the vanilla side (54.5 6 37.5). Chicks not exposed to strawberry did not differ [t(10) 5 1.26, NS] in the time they spent on the strawberry (mean 6 SD 76.8 6 34.7) or vanilla side of the cage (103.2 6 34.7). “Gustatory” preference test. There was a significant difference [t(8) 5 2.875, p 5 0.0207] between the percentage amount of strawberry-flavoured and vanilla-flavoured water drunk by the chicks. Chicks exposed to strawberry before hatching drank more strawberry-flavoured water (mean 6 SD 78 6 13.8) than those not exposed to strawberry before birth (45.7 6 20). Paired t-tests to examine the percentage amount of strawberry-flavoured and vanilla-flavoured water drunk by each group were then undertaken. Chicks exposed to strawberry before hatching drank significantly [t(4) 5 4.523, p 5 0.0106] more strawberryflavoured water (mean 6 SD 78 6 13.8) than vanilla-flavoured water (22 6 13.8). Chicks not exposed to strawberry did not differ [t(4) 5 0.46, NS] in the amount of strawberry-flavoured water (mean 6 SD 45.7 6 20) or vanilla-flavoured water they drank (54.3 6 20).
Procedure Two choice tests were undertaken in this study. First, 26 chicks from each group were examined in the “olfactory” and “gustatory” tests and given the choice between vanilla or water. Both the “olfactory” preference test (Day 2 posthatch) and the “gustatory” test (Day 4 posthatch) were performed identically to the tests described above in Experiment 1. The exception to this was that
SUMMARY
Exposure to strawberry before hatching results in a preference that did not affect the chick’s response to a novel, vanilla in this case, chemosensory stimulus. Further when given a preference test between strawberry and vanilla, only those chicks exposed to strawberry before hatching showed a preference for strawberry.
CHICKEN EMBRYONIC LEARNING These results indicate that the preference acquired by the chicks to the strawberry in Experiment 1 would appear to be specific to stimulus experienced before hatching. EXPERIMENT 4 One problem with the procedure adopted to rear the chicks in Experiment 1 is that the chicks may experience the stimulus between the time of hatching and the time of testing. Exposure to the stimulus stopped on Day 20 to prevent this possibility. However, although the strawberry solution was not applied on Day 21, traces may still have remained from previous treatments, especially in the groups exposed to the stimulus by rubbing it on their shells or by injection into the egg. It is possible in these groups that the chicks could have acquired the preference for strawberry in the period between hatching and removal from the incubator from smelling the strawberry on the shell or by eating the shell either during or after hatching. To rule out this possibility, a further study was undertaken in which eggs at Day 21 were artificially hatched and removed to a new cage and room, preventing exposure to the stimulus after hatching. Subjects Treatment conditions were the same as reported above for Groups 2 (n 5 22) and 4 (n 5 24) of Experiment 1. The only exception to the treatment undertaken by these eggs was that on Day 21 eggs were closely watched. When cracks in the shell were noted, but prior to the appearance of a hole, the egg was removed from the incubator, the shell was opened, and the chick was hatched and placed in a new cage. This ensured the chick had no experience of the strawberry stimulus once hatched prior to testing. Procedure Both the “olfactory” preference test (Day 2 posthatch) and the “gustatory” preference test (Day 4 posthatch) were performed identically as described in Experiment 1. RESULTS
“Olfactory” Preference Test There was a significant difference [t(22) 5 5.01, p 5 0.001] between the time (seconds) spent on the strawberry side of the maze by pairs of chicks exposed before hatching to strawberry (mean 6 SD 71.3 6 38.5) and those not so exposed (10.8 6 13.2). “Gustatory” Preference Test For this test there were five groups of four chicks from the strawberry exposure group and six from the control, no exposure, condition. The results indicated there was a significant difference [t(9) 5 3.209, p 5 0.011] between the percentage amount of strawberry water drunk by chicks exposed before hatching to strawberry (mean 6 SD 21.6 6 13.6) and those not so exposed (3.7 6 2.3). SUMMARY
In both “olfactory” and “gustatory” tasks chicks exposed to strawberry before hatching showed a greater preference for this stimulus than chicks not so exposed. Thus exposure to a chemosensory stimulus before hatching alone is sufficient to change the preferences of chicks after hatching.
137 DISCUSSION Previous studies have indicated the chicken is able to learn auditory stimuli before hatching. The results of this study extend these findings by demonstrating that the chicken is able to learn chemosensory stimuli before hatching. Experimental exposure to the stimuli in this study took place over 5 days before hatching. However, it is possible that in the shell and injection groups the embryo could have experienced the stimulus on Day 21 from its continued presence on the shell or inside the egg. Thus it is reasonable to assume that at some point during this time (E15–E21, E 5 embryonic day) the chicken embryo’s olfactory and/or gustatory senses commence functioning. Previous studies have demonstrated a functioning chemosensory sense at E21 (29). Our results from the air exposure group, where exposure was confined to E15–E20, suggest this may be functioning even earlier. The olfactory system begins its development early in chick embryogenesis (1,20). Outgrowths of olfactory neurons are seen at E3 and olfactory bulb morphogenesis begins at E7. Around this time proteins involved in the reception and transduction of olfactory signals are found. The olfactory epithelium is mature by E19. It is most likely that the peripheral olfactory system is mature prior to central processing of olfactory signals although exactly when the olfactory system becomes functional is unknown. The functional ontogenesis of the gustatory system is less well understood. Taste buds develop rapidly from E16 (no buds) to hatching (7) where numbers approximate half that of the adult bird. Again when they become functional is unknown. The behavioural evidence presented in this study suggests both olfactory and gustatory receptors are functional before hatching. Studies of the embryonic learning of chemosensory stimuli find it difficult to separate the roles of gustation and olfaction in exposure and the resulting preference. Two preference tests were therefore conducted in an attempt to separate the roles of smell and taste in the exhibition of the preference after hatching. Chicks exposed to strawberry before hatching exhibited a change in preference to this when tested in the “olfactory” task and in the “gustatory” task. There was undoubtedly some olfactory stimulus present in the “gustatory” test which could have guided the chick to the bottle. However, a gustatory component was most certainly involved once the chick began to drink and gustation was probably involved in the continuation of drinking following the initial consumption of the fluid. Thus the “gustatory” test may have contained some olfactory stimuli as well as gustatory stimuli. Taste stimuli played very little role in the “olfactory” test. Although chicks walked on the shavings, there was little observed pecking of the shavings by the chicks and hence little opportunity for gustatory stimuli in this preference. However, the exact role of odour and/or taste in the preferences observed has yet to be determined. At the exposure stage both taste and smell may have been involved. The chick begins to swallow amniotic fluid from about halfway through incubation (21) and possibly earlier (17). The nares are blocked by tissue for most of the incubation period; this degenerates toward the end of this period and the nares are open by E20 (29). Whilst this may restrict the access of stimuli directly to the olfactory receptors, the swallowing activities of the embryo may stimulate the olfactory receptors by moving fluid from the mouth to the nasal cavities. Such fluid movement is observed in the human fetus from midway through gestation (P.G.H., personal observation) and may occur in the chicken embryo. The gustatory sense however may have greater direct access to chemosensory stimuli. It should be noted that
138
SNEDDON, HADDEN AND HEPPER
exposure before hatching resulted in both an “olfactory” and a “gustatory” preference. If no olfactory exposure occurred before hatching, this would indicate intermodal transfer of information between gustatory and olfactory senses. Olfactory processes in birds have often been viewed as less important than other sensory processes, especially auditory, and have by comparison been little studied. However, recent years have seen renewed interest in avian olfaction. Examination of olfactory thresholds reveals birds (passerines) have detection thresholds similar to those of rats (5). Olfaction has been postulated to play a role in homing (2,33). The role of olfaction in feeding behaviour has aroused considerable interest (19,30). The studies reported here indicate that birds have a functioning chemosensory system as embryos and suggest a role for chemosensation in the perinatal period. The methods of exposure used in the study aimed to mimic any “natural” way that the chicken embryo may have contact with chemosensory stimuli. Whilst the direct injection of stimuli into the egg was purely experimental, exposure to stimuli from the air surrounding the egg or from its presence on the egg shell may both occur naturally during the embryo’s time in the egg. What function the learning of chemosensory stimuli before hatching may have for the chick is unknown. In other species prenatal chemosensory exposure may influence subsequent recognition and/or food preferences (9) and it is possible it may serve similar functions in birds. Visual imprinting and auditory imprinting have been well established in newly hatched chicks (13), and there is some evidence for olfactory imprinting in chicks (31). In the embryo there is much evidence for auditory imprinting (8,14,18). Rogers has speculated that the chicken embryo may imprint on the odour of its surroundings, e.g., from the nest or hen (20). It is possible that odours from the nest or the mother when incubating her eggs would pass across the shell and stimulate the chemosensory receptors of the chicken embryo.
Indeed such stimuli would have access to the embryo from well before hatching and this may increase the concentration of the chemosensory stimulus in the egg. Hence, if the olfactory or gustatory system is functioning before hatching, the possibility of embryonic olfactory imprinting for maternal or nest recognition should be considered. Six-day-old chicks show a preference for their own soiled nest litter over strange or clean bedding (16). However, given the 6 days of exposure after hatching, this preference cannot be attributed to embryonic exposure. This study adds to the growing list of species in which prenatal chemosensory learning has been established. Olfaction is regarded, in evolutionary terms, as the oldest sense and its functioning across different species in the perinatal period shows remarkable similarity and universality. The finding of “prenatal” chemosensory learning across major animal groups may suggest some evolutionary importance to this ability perhaps ensuring recognition of parent(s), kin, nestmates, or the “nest site” or influencing feeding behaviour (11). Exposure of chicken embryos to a chemosensory stimulus affects their subsequent preference for this stimulus after birth. The change in preference is specific to the stimuli experienced before hatching. The chicken embryo may experience chemosensory stimuli naturally during the course of incubation, and chemosensory learning may have some function, possibly enabling recognition in the newly hatched chick. Studies are currently underway to examine the function of chemosensory stimuli in the perinatal period in the chick. ACKNOWLEDGEMENTS
We acknowledge the support of Prof. K Brown, School of Psychology, for providing facilities and Ken McCarroll for husbandry support. All studies were performed under license issued by Department of Health and Social Services (DHSS). The comments and suggestions of the reviewers have greatly improved this paper.
REFERENCES 1. Ayer-Le Lie`vre, C.; Lapointe, F.; Leibovici, M. Avian olfactory neurogenesis. Biol. Cell. 84:25–34; 1995. 2. Bingman, V. P.; Benvenuti, S. Olfaction and homing ability of pigeons in the Southeastern United States. J. Exp. Zool. 276:186 –192; 1996. 3. Board, R. G. Properties of avian egg shells and their adaptive value. Biol. Rev. 57:1–28; 1982. 4. Caubet, Y.; Jaisson, P.; Lenoir, A. Preimaginal induction of adult behaviour in insects. Q. J. Exp. Psychol. 44B:165–178; 1992. 5. Clark, L.; Avilova, K. V.; Bean, N. J. Odor thresholds in passerines. Comp. Biochem. Pysiol. A Physiol. 104:305–312; 1993. 6. Courtenay, S. C. Learning and memory of chemosensory stimuli by underyearling coho salmon Oncorhynchus kisutch (Walbaum). Vancouver, BC: University of British Columbia, 1989 (PhD Thesis). 7. Ganchrow, J. R.; Ganchrow, D. Development of taste buds in the chick-embryo (Gallus gallus domesticus). Ann. NY Acad. Sci. 510: 307–308; 1987. 8. Gottlieb, G. Development of species identification in ducklings: IX. The necessity of experiencing normal variations in embryonic auditory stimulation. Dev. Psychobiol. 15:507–517; 1982. 9. Hepper, P. G. Adaptive fetal learning: Prenatal exposure to garlic affects postnatal preferences. Anim. Behav. 36:935–936; 1988. 10. Hepper, P. G. Human fetal “olfactory” learning. Int. J. Prenatal Perinatal Psycho. Med. 7:147–151; 1995. 11. Hepper, P. G. Fetal memory: Does it exist? What does it do? Acta Paediatr. Suppl. 416:16 –20; 1996. 12. Hepper, P. G.; Waldman, B. Embryonic olfactory learning in frogs. Q. J. Exp. Psychol. 44B:179 –197; 1992. 13. Horn, G. Imprinting, memory and the brain. Oxford: Clarendon Press; 1985.
14. Impekoven, M. Responses of laughing gull chicks (Larus atricilla) to parental attraction and alarm calls, and effects of prenatal auditory experience on the responsiveness to such calls. Behaviour. 61:250 – 278; 1976. 15. Isingrini, M.; Lenoir, A.; Jaisson, P. Pre-imaginal learning as a basis of colony-brood recognition in the ant, Cataglyphis cursor. Proc. Natl. Acad. Sci. USA. 82:8545– 8547; 1985. 16. Jones, R. B.; Gentle, M. J. Olfaction and behavioural modifications in domestic chicks (Gallus domesticus). Physiol. Behav. 43:917–924; 1985. 17. Kuo, Z. Y. Ontogeny of embryonic behaviour in Aves. I. The chronology and general nature of behaviour in the chick embryo. J. Exp. Zool. 61:395– 430; 1932. 18. Lickliter, R.; Stoumbos, J. Modification of prenatal auditory experience alters postnatal auditory preferences of bobwhite quail chicks. Q. J. Exp. Psychol. 44B:199 –214; 1992. 19. Marples, N. M.; Roper, T. J. Effects of novel colour and smell on the response of naive chicks towards food and water. Anim. Behav. 51:1417–1424; 1996. 20. Rogers, L. J. The development of brain and behaviour in the chicken. Cambridge: CAB International; 1995. 21. Romanoff, A. L. The avian embryo. New York: Macmillan; 1960. 22. Schaal, B.; Orgeur, P. Olfaction in utero: Can the rodent model be generalized? Q. J. Exp. Psychol 44B:245–278; 1992. 23. Sedlacek, J. Functional characteristics of the centre of the unconditioned reflex in elaboration of a temporary connection in chick embryos. Physiol. Bohemoslov. 11:313–318; 1962. 24. Shindler, K. M. A three year study of fetal auditory imprinting. J. Wash. Acad. Sci. 74:121–124; 1984.
CHICKEN EMBRYONIC LEARNING 25. Smotherman, W. P. In utero chemosensory experience alters taste preferences and corticosterone responsiveness. Behav. Neural Biol. 36:61– 68; 1982. 26. Smotherman, W. P. Odor aversion learning by the rat fetus. Physiol. Behav. 29:769 –771; 1982. 27. Smotherman, W. P.; Robinson, S. R. The rat fetus in its environment. Behavioral adjustments to novel, familiar, aversive and conditioned stimuli presented in utero. Behav. Neurosci. 99:521–530; 1985. 28. Smotherman, W. P.; Robinson, S. R. Habituation in the rat fetus. Q. J. Exp. Psychol. 44B:215–230; 1992.
139 29. Tolhurst, B. E.; Vince, M. A. Sensitivity to odours in the embryo of the domestic fowl. Anim. Behav. 24:772–779; 1976. 30. Turro, I.; Porter, R. H.; Picard, M. Olfactory cues mediate food selection by young chicks. Physiol. Behav. 55:761–767; 1994. 31. Vallortigara, G.; Andrew, R. J. Olfactory lateralization in the chick. Neuropsychologia. 32:417– 423; 1994. 32. Vince, M. A. Taste sensitivity in the embryo of the domestic fowl. Anim. Behav. 25:797– 805; 1977. 33. Wiltschko, R. The function of olfactory input in pigeon orientation. Does it provide navigational information or play another role? J. Exp. Biol. 199:113–119; 1996.