- Email: [email protected]
Operant, open-field and tonic immobility behaviours in chickens with forebrain injections of cycloheximide or glutamate
Behavioural Brain Research, 4 (1982) 19-32
19
Elsevier Biomedical Press
OPERANT, OPEN-FIELD AND TONIC IMMOBILITY BEHAVIOURS IN CHICKENS WITH FOREBRAIN INJECTIONS OF CYCLOHEXIMIDE OR G L U T A M A T E
P A U L R. S A N B E R G , I A N J. F A U L K S * , J U D I T H M. A N S O N and R I C H A R D F. M A R K
Department of Behavioural Biology, Research School of Biological Sciences, Australian National University, Canberra, ACT 2600 (Australia) (Received M a r c h 12th, 1981) (Accepted April 7th, 1981)
Key words: cycloheximide - glutamate - operant behavior - open-field - tonic immobility retarded learning - emotional reactions - fear - chicken
SUMMARY
Chickens that had received bilateral injections of cycloheximide or glutamate into the forebrain on day 2 of life and tested 4 weeks later showed no deficit in acquisition, performance or extinction of continuously reinforced appetitive key-pecking as compared to control birds injected with saline. However, chickens that had received injections of cycloheximide and were subsequently tested in an open-field apparatus took longer to leave the first square, defaecated more, and pecked, preened and moved about less than controls. They also showed longer durations of tonic immobility. Those injected with glutamate exhibited similar behaviour but were not significantly different from controls in the open field latency to leave the first square, defaecation, or tonic immobility tests. The above treatments have previously been described as producing permanent slowed learning in chickens on a pebble-floor task. Our results suggest that learning mechanisms may not be disrupted as shown by normal performance in a simple operant task but that enhanced emotionality or fear of novelty as revealed in the open field tests may interfere with the expression of learning behaviour in some situations.
* To w h o m all correspondence should be addressed. 0166-4328/82/0000-0000/$02.75 © Elsevier Biomedical Press
20 INTRODUCTION
In 1972 Rogers and Mark [14] reported that intracranial injections of cycloheximide (a cytoplasmic ribosomal protein synthesis inhibitor), given during the first week after hatching, would subsequently render the animal incapable of learning at the normal rate on a pebbled-floor task. Chicks injected with cycloheximide failed to reach the control levels of performance, that is they failed to choose predominantly grain, in preference to pebbles, in the latter half of the 60 pecks of the test. Sensory input (visual and auditory) within 3 h following the injections were required to slow subsequent acquisition of these behavioural tasks. Cycloheximide-treated birds were also slower to habituate to visual and auditory stimuli. In addition, Rogers and Anson [12] have shown that birds treated with cycloheximide persisted in pecking at preferred food more than saline-injected control animals. Rogers and her colleagues [12, 15] have found the injection parameters of dose, volume and age of chicks to be critical in obtaining all these effects. Hambley and Rogers [6] have postulated that cycloheximide produced these long-term behavioural changes in chicks by the decreased utilization for protein synthesis and consequent accumulation of the amino acids, glutamate and aspartate. Forebrain injections of these neurotransmitters, particularly glutamate, were shown to produce behavioural deficits on the pebble floor task qualitatively similar to those seen in the cycloheximide-injected birds [6, 23]. The factors that contribute to performance on the pebble floor are not clearly understood [10, 11]. Therefore, the present study was conducted in order to directly compare the long-term effects of neonatal injections of glutamate and cycloheximide on a variety of behavioural tasks conventionally used to measure changes in learning (appetitive key-pecking), activity (open-field) and emotionality (open-field and tonic immobility) in birds. A preliminary report has been presented elsewhere [22]. MATERIALS AND METHODS
(A) Material Birds and Housing White leghorn × Black Australorp male chickens were obtained from Research Poultry Farm (Victoria) on day 1 of life. They were housed in groups of 4-6 birds until day 14 when they were housed in pairs for the remainder of the study in 23 × 23 × 29 cm metal cages with a clear perspex front. Warmth (25 °C) and light were provided continuously by overhead lamps (25 W). Chicks had free access to grain (Hutmill, Victoria) and water (tap water with terramycin added) during the study, except when under deprivation schedules as described below.
21
Drug administration On day 2 of life, the chicks were randomly assigned to 3 groups and given bilateral intracerebral injections of either cycloheximide (20 /~g/hemisphere), glutamate (210 #g/hemisphere) or saline vehicle. The glutamate solution was buffered to pH 7.2 with Na2PO4. All injections were given in a vehicle of 25/~1 saline and performed freehand into the middle of each forebrain using a Hamilton 50/zl syringe. A 26.5-gauge needle was used and fitted with a rubber stopper to prevent penetration beyond 3 mm below the surface of the skull. These doses and volumes were used since they were found to provide the optimal conditions for producing pebble floor deficits [12].
Appetitive key-pecking Four standard BRS Foringer Skinner boxes for pigeons, enclosed in soundproof chambers, were used. Each was fitted with a centre illuminated pecking key located immediately above a grain feeder. They were modified with a grid floor of adjustable height, enlarged grain access hole and placement of a photocell beam across the food cubicle so that the beam was interrupted when the animal's head entered, as described elsewhere [16]. Programming and stimulus delivery were controlled by Digital K-logic solid state circuitry. Keypecks were recorded on printing counters converted from Royal 310P printing calculators. Photocell interruptions and reinforcements were recorded on counters converted from National Semiconductor 850A calculators [18]. Chicken starter crumble was used as the reinforcer.
Open-field A 80 x 100 cm platform with 50 cm high walls was used as the open-field. It was painted white and the floor was divided by black lines into 20 squares of 20 cm2. The open-field was placed in a 3.1 x 6.5 m laboratory with fluorescent lighting 2.8 m above the floor.
Tonic immobility Tonic immobility was measured on the floor of a BRS chamber with the door open for visual observation. (B) Methods
Acquisition and extinction of keyopecking At three weeks of age, 16 cycloheximide-, 10 glutamate- and 14 salineinjected chickens were restricted to 1.5 h of feeding daily sometime between 21.00 and 23.00 h. One week later the animals were given a 20 min magazine training session on both days 1 and 2. The animals were gently placed in the operant chamber, which had the grain feeder up and extra grain in the food
22 hopper (days 1 and 2) and scattered on a paper towel on the grid floor immediately before the hopper entrance (day 1). On days 3-5, the animals were given hopper training over 20 min sessions. On days 3 and 4 the feeder (hopper) delivered grain to the food cubicle for 10 sec followed by 5 sec of non-delivery. On day 5 food was delivered for 10 sec, followed by 20 sec of non-delivery. Photocell interruptions were measured over days 1-5. Days 6-10 consisted of autoshape-continuous reinforcement (CRF) training. The animals were given daily sessions of 30 min each, in which the hopper delivered grain for 5 sec every 30 sec (days 6 and 7) or 60 sec (days 8, 9 and 10). In each session the free grain delivery was preceded by 5 sec of red illumination of the pecking key, which ceased when the grain was delivered (autoshape). In addition, the chicken received 5 sec access to grain (CRF), if the key was pecked when not illuminated. A printing counter recorded those pecks which occurred during either the autoshape or CRF schedules over time (30 sec periods for days 6 and 7; 60 sec periods for days 8, 9 and 10). On days 11-19 the chickens were given CRF performance sessions only. These were 30 min long; the pecking key was illuminated red continuously, and only grain (5 sec access) contingent on key-pecking was available. On days 20-24 reinforcement was discontinued. Each of these 5 extinction sessions lasted until the chickens ceased to respond for 3 consecutive minutes. Both latency and number of responses to the extinction criterion were recorded.
Open-field At 4 weeks of age, 10 cycloheximide-, 10 glutamate- and 9 saline-injected chickens were placed individually in the central square of the open-field. For the next 5 min the following behaviours were observed: (1) the latency (sec) to leave the first square; (2) the number of squares entered by both feet (ambulation); (3) the number of pecks at the floor; (4) the number of preening episodes; and (5) the number of times the animals defaecated. The open field was cleaned and wiped with a weak vinegar solution between each test.
Tonic immobility At 4 weeks of age, 19 cycloheximide-, 17 glutamate- and 20 saline-injected chickens were tested individually for tonic immobility. The bird was held down by hand on its right side for 15 sec. Tonic immobility was measured as the time in sec from when the animal was released by the experimenter until it righted itself and rose to its feet. Additional induction attempts were given 60 sec after each unsuccessful attempt to elicit tonic immobility. If the animal failed to show tonic immobility over 5 attempts it was given a score of 0 sec. The floor was cleaned and wiped with a weak vinegar solution between each test. Over all behavioural experiments, treatment and testing were completely randomized.
23
Statistics One-way analyses of variance were used on the appetitive key-pecking acquisition data, open-field and tonic immobility data. Bonferoni [7] multiple comparison tests were used to test significance between groups. Analysis of variance for trends of trial means [2] were performed on the appetitive keypecking acquisition and extinction data. Because of skewness and heterogeneity of variance on magazine and hopper training, and open field latencies, nonparametric Mann-Whitney U-tests were used to test significance between groups. Two-tailed significance tests were used throughout the analyses. RESULTS
Appetitive key-pecking The median photocell counts for magazine and hopper training are shown in Fig. 1. Only on day 1 was there any significant effect. The birds treated with cycloheximide broke the photocell beam in the food cubicle less often than birds treated with saline (U = 61, nl/n2 = 14/16, P < 0.05). The performance of birds treated with glutamate and those given cycloheximide (U = 73, nl/n2 = 14/16, P < 0.05). The performance of birds treated with glutamate and those given cycloheximide 0d = 73, nl/n2 = 10/16, P >0.05) on this measure was not statistically different, although visual observation suggested that the birds MAGAZINE TRAINING DAY I
C
DAY 2
HOPPER TRAINING DAY 3
DAY 4
DAY 5
500
o U 400 I
U
•~ 300
e'Q,.
•
j
._~ 200 "(3
•
loo SAL CXM GLU
SAL CXM GLU
SAL CXM GLU
SAL CXMGLU
$AL CXMGLU
Fig. 1. Median responses of head entries in food cubicle during acquisition of magazine and hopper training. SAL, saline group; CXM, cycloheximide group; CLU, glutamate group.
24 9O
AUTOSHAPE
CRF
80 C
o
~
zo
u
60
@
~
o
//A
//A I //A //A //A //A //A //A //A
5O
\\~
4O
\\4 \
"5
\\Xl \\~ \\N
/ /A
\\'t
//A
\\N
30 //A zz A ///I
tO
/~A
20'
\-,
" "xl "4 \ \ ~ \\N
//11
//A //A
¢1
10
SAL
CXM
SAL
GLU
CXM
GLU
Fig. 2. Mean time and standard error of the mean to first keypeck response in the autoshape or continuous reinforcement (CRF) periods during acquisition of continuously reinforced key-pecking.
400
AUTOSHAPE-C R F
C RF
o--o SAUNE a--a CYCLOHEXIMIDE a--~ GLUTAMATE
350
c 0 ev
300 250
~.
200
~-
150
v O
O
100
I
I
I
I
I
I
i
I
[
I
1
I
I
I
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Meanresponsesduringacquisitionofcontinuouslyreinforced Days
Fig. 3. key-pecking. Autoshape-CRF means the session days in which autoshaping procedures were used.
25 treated with cycloheximide ate much less grain on day 1 than either of the other groups. The results of the autoshape-CRF are shown in Figs. 2 and 3. The mean latencies in total min to the first key-peek (Fig. 2) during the red illumination (autoshape) or non-illumination (CRF) periods were not significantly different between groups (F = 1.59, df = 3.37, P > 0.05, and F < 1, respectively). There were no differences between the mean response rates of the different groups (Fig. 3) (F < 1). All 3 groups showed a significant increase in the number of mean key-peck responses over the 13 days the birds were tested ( F = 52.4, d f = 13,481, P < 0.001). On any one day the mean key-peck response did not CRF E X T I N C T I O N 300['--I
cO t,. t,_
U
200-
I
SALINE CYCLOHEXIMIDE GLUTAMATE
o u G)
v
100-
Y/, Y/,
tO
0 4035 c
o
30
i,_ t,_
k.)
25
o 20 E tO
15 10
-I
5 0 2
3
4
5
Days
Fig. 4. Mean responses and standard error of the mean (upper panel) and mean latency (see) to criterion and standard error of the mean latency (lower panel) during extinction of continuously reinforced key-pecking.
26 differ significantly between the 3 groups (e.g. Day 19, F < 1). However, all 3 groups showed a significant decrease in key-pecking within each trial (e.g. Day 19, F = 15.79, df = 14,518, P < 0.001). Extinction of the key-pecking response is shown in Figs. 4 and 5. There were no significant differences between the mean time and the mean number of responses the groups took to reach the extinction criterion (Fig. 4). Analysis of variance of the key-pecking response revealed no significant effects of drug treatment (F = 1.42, df = 2,37, P > 0.05) or interaction of drug treatment with days ( F = 1.33, d f = 8,148, P >0.05). A significant effect over days, indicative of extinction behaviour, was noted ( F = 98.73, df = 4,148, P < 0.001). Analysis of variance of the latencies also revealed a non-significant effect of drug treatment (F = 1.06, df = 2,37, P >0.05) and a significant effect over session days ( F = 241.75, df = 4,148, P < 0.001). There were no significant differences
30
C R F EXTINCTION 25
20
o--o
o--a =--=
SALINE CYCLOHEXIMIDE GLUTAMATE
5 10
5
10
15
20
2.5
30
55
TriaJs
Fig. 5. M e a n responses over trials (1 trial = 60 sec) during extinction days 1 and 5 of continuously reinforced key-pecking.
27 between groups on each extinction day over time (Ps >0.05). The results of extinction sessions on days 1 and 5 with time are shown in Fig. 5.
Open-field Figure 6 shows that the birds treated with cycloheximide took significantly longer to leave the first square than birds treated with saline (U = 17, nl/n2 = 10/9, P <0.05), however, birds given glutamate did not differ from cycloheximide- or saline-treated animals (U = 48, nl/n: = 10/10, and U = 28, n~/n2 = 10/9 respectively, Ps >0.05). There was a significant effect of groups on the ambulation measure ( F = 5.3, df = 2,26, P < 0.02). Multiple comparison tests revealed that both the cycloheximide- ( F = 9 . 1 4 , d f = 1,26, P <0.02) and glutamate-treated ( F = 6.91, d f = 1,26, P < 0.05) groups passed significantly less squares than did controls. There was no significant difference between cycloheximide or glutamate scores (F < 1) (Fig. 6). Additional open-field data is shown in Fig. 7. There were significant group effects for the other behavioural measures (Fig. 7). Multiple comparison tests revealed that both the cycloheximide-treated and glutamate-treated groups pecked at the floor (Fs = 20.49 and 10.15, df = 1,26, Ps < 0.0005 and 0.02, respectively) and preened (Fs = 7.32
280
LATENCY TO LEAVE FIRST SQUARE
I
260
v~ 240
i
- 0¢..
o
~
220
¢,.
200
180
160
AMBULATION
16
O 12
Z c
3 o 8
.Q cQ
4
Q
2
140
o
0 SAL
CXM
GLU
SAI
CXM
GLU
Fig. 6. Latency to leave first square and ambulation in saline (SAL)-, cycloheximide (CXM)- and glutamate (GLU)-injected chicks tested for 5 min in an open-field. Bars represent standard errors of the means.
28
PECKS AT GROUND
4.0
PREENING -EPISODES
DEFAECATION
3.5
e-
3.0
> LU
,_,_ o
2.5
..a E
2.0
Z e.-
1.5
0
~
1.o 0.5
SAL
CXM
GLU
SAL
CXM
GLU
SAL
CXM
GLU
Fig. 7. Number of pecks at ground, preening episodes and faecal boli in saline (SAL)-, cycloheximide (CXM)- and glutamate (GLU)-injected chicks tested for 5 min in an open-field. Bars represent standard errors of the means.
and 10.81, d f = 1,26, Ps < 0.05 and 0.01, respectively) less than saline-treated birds. On the other hand, birds treated with cycloheximide and glutamate tended to defaecate more than saline-treated birds although the scores did not quite reach significance (Fs =4.48 and 6.06, df--1,26, Ps <0.13 and 0.06, respectively). There were no differences between birds treated with cycloheximide and glutamate in these three behavioural measures.
Tonic immobility Analysis of variance of the tonic immobility results (Fig. 8) showed a significant group effect ( F = 3.42, df--2,53, P < 0.05). Multiple comparison tests revealed a significant difference between cycloheximide- and saline-injected birds ( F = 6.57, df = 1,53, P < 0.05). Birds injected with glutamate were not significantly different from either cycloheximide- (F = 2.93, df = 1,53, P > 0.05) or saline-treated birds (F < 1). DISCUSSION
Neonatal injections ofcycloheximide or glutamate did not alter either rate of acquisition or asymptotic performance of key-pecking behaviour in chickens.
29 TONIC IMMOBILITY 800 700 600
500 U 4)
,- 400 3O0 200 100
m
SAL
CXM
GLU
Fig. 8. Duration of tonic immobility in saline (SAL)-, cycloheximide (CXM)- and glutamate (GLU)-injected chicks. Bars represent standard errors of the means.
Regardless of treatment, all groups performed similarly throughout hopper, autoshaping and CRF training procedures. Furthermore, there were no differences between groups on extinction of the CRF schedule. Only on the first day of magazine training did the group treated with cycloheximide respond less than the other groups. This difference probably reflects an enhanced fear response to novel situations, since chickens treated with cycloheximide show enhanced tonic immobility. Tonic immobility is a reliable measure of emotionality and fear in chickens [3, 4] and has been shown to be negatively correlated with depressed magazine training [19]. Tonic immobility is also inversely proportional to the position of each bird in the peck-order [3]. Chickens treated with cycloheximide fall to the bottom of the peck-order [15] and display altered levels of emotionality when reacting to visual and auditory stimuli (Anson, unpublished data). Anson (unpublished data) also found a pattern of responding consistent with enhanced emotionality in cycloheximide treated birds tested on the pebble floor task; that is, they took significantly longer to make the first 5 pecks than birds treated with saline. Birds treated with cycloheximide or glutamate were less active in an open-field situation than controls as reflected by fewer ambulatory, pecking and preening episodes. This relative immobility may indicate increased fear and general emotionality, since both groups tended to defaecate more than salineinjected birds. Some investigators have questioned the use of defaecation as a
30 measure of emotionality [1, 25], while the latency to leave the first square is considered a better measure [9]. Birds treated with cycloheximide, but not glutamate, showed significantly longer latencies to leave the first square than did controls. A similar pattern of results was also found in the tonic immobility experiment. The differentiation between the cycloheximide and glutamate effects on the two emotionality measures of open-field latency and tonic immobility, compared to those induced by saline, could mean that the long-term effects of these drugs are produced by different mechanisms. The fact that the differences between glutamate and cycloheximide over all behavioural tests were not significant may imply that the effects of glutamate injections alone are simply not as strong as cycloheximide-induced effects. Cycloheximide injections increase the brain levels of aspartate and other amino acids in addition to that of glutamate [6]. Previous studies with the pebble floor task [6, 12-15, 23] have shown that birds injected with cycloheximide or glutamate require many more pecks than saline-injected chicks to reach a similar level of performance. These results were interpreted as indicating that cycloheximide and glutamate interfered with associative and/or memory processes to produce slowed learning. In the present study, treatment with cycloheximide or glutamate had no significant effect on the rate of learning of an appetitive instrumental response. Perhaps alterations in pebble floor performance after treatment with cycloheximide or glutamate reflect not a learning deficit but altered emotionality. The results from the open-field study and the tonic immobility experiment suggest this could be the case. However, changes in the general emotional state of an animal alone cannot easily explain why [13, 23] perceptual input for the 3 h post-injection period is necessary for the drugs to alter subsequent behaviour on the pebble floor learning task. More work is needed before firm conclusions can be made about what effects cycloheximide and glutamate are having on the brains and behaviour of young chickens. Originally, Rogers et al. [15] reported that cycloheximide injections did not produce light or electron microscopic evidence of cellular damage in the hippocampal and Wulst areas. However, since glutamate is a known neurotoxic agent, especially in neonatal animals [5, 8] more detailed studies in other areas need to be carried out to elucidate the potential histological effects of cycloheximide and glutamate forebrain injections in young chickens. Previous research has shown permanent changes in neurotransmitter pharmacology following injection of kainate, an analogue of glutamate, into the rat striatum [17, 21] resulting in permanent changes in activity and emotionality [20]. The known role of monoaminergic systems in emotionality [24] suggest that studies investigating permanent changes in this transmitter system following cycloheximide or glutamate injections may be fruitful.
31 ACKNOWLEDGEMENTS
The authors appreciate the excellent assistance of Drs. K. Gillette, J. Irwin, W.P. BeUingham, I.G. Morgan and K.E. Sanberg throughout the course of this study. This study was completed in partial fulfilment of the Doctor of Philosophy degree from the Australian National University to P.R. Sanberg. REFERENCES 1 Archer, J., Tests for emotionality in rats and mice: a review, Anita. Behav., 21 (1973) 205-235. 2 Edwards, A.L., Experimental Design in Psychological Research, Holt, Rhinehart and Winston, New York, 1972, pp. 334-338. 3 Gallup, G.G., Jr., Tonic immobility as a measure of fear in domestic fowl, Anim. Behav., 27 (1979). 4 Gallup, G.G., Jr., Ledbetter, D.H. and Maser, J.D., Strain differences among chickens in tonic immobility: evidence for an emotionality component, J. comp. physiol. Psychol., 90 (1976) 1075-1081. 5 Garattini, S., Evaluation of the neurotoxin effects of glutamic acid. In R.J. Wurtman and J.J. Wurtman (Eds.), Nutrition and Brain, Vol. 4, Raven Press, New York, 1979, pp. 79-124. 6 Hambley, J.W. and Rogers, L.J., Retarded learning induced by intracerebral administration of amino acids in the neonatal chick, Neuroscience, 4 (1979) 677-684. 7 Miller, R.G., Simultaneous Statistical Influence, McGraw-Hill, New York, 1966. 8 Olney, J.W., Brain damage and oral intake of certain amino acids, Advanc. exp. Med. Biol., 69 (1976) 497-506. 9 Ossenkopp, K.-P., The open-field test as a rapid and sensitive behavioural measure in white peking ducklings Anas platyrhynchos, Bird Behav., 2 (1980) 23-35. 10 Reymond, E. and Rogers, L.J., Deprivation of the visual and tactile aspects of food important to learning performance of an appetitive task by chicks, Behav. Neural Biol., 31 (1981) 425-434. 11 Reymond, E. and Rogers, L.J., Diurnal variation in learning performance of young chicks, Anim. Behav., 29 (1981) 241-248. 12 Rogers, L.J. and Anson, J.M., Cycloheximide produces attentional persistence and slowed learning in chickens, Pharmacol. Biochem. Behav., 9 (1978) 735-740. 13 Rogers, L.J. and Drennen, H.D., Cycloheximide interacts with visual input to produce permanent slowing of visual learning in chickens, Brain Res., 158 (1978) 479-482. 14 Rogers, L.J. and Mark, R.F., Retarded learning in chicks after brief inhibition of brain protein synthesis, Proc. Aust. Physiol. Pharmacol. Soc., 3 (1972) 119. 15 Rogers, L.J., Drennen, H.D. and Mark, R.F., Inhibition of memory formation in the imprinting period: irreversible action of cycloheximide in young chickens, Brain Res., 79 (1974) 213-233. 16 Sanberg, P.R., Measuring feeding responses in operant research, Psychol. Rep., 45 (1979) 942. 17 Sanberg, P.R., Haloperidol-induced catalepsy is mediated by post-synaptic dopamine receptors, Nature (Lond.), 284 (1980) 472-473. 18 Sanberg, P.R. and Bellingham, W.P., Digital counters: inexpensive alternatives, Physiol. Behav., 23 (1979) 795-797. 19 Sanberg, P.R., Faulks, I.J., Bellingham, W.P. and Mark, R.F., Relationship between tonic immobility and operant conditioning in chickens Gallus gallus, Bird Behav., 3 (1981) 51-56. 20 Sanberg, P.R., Pisa, M. and Fibiger, H.C., Locomotor activity, exploration and neophobia in rats with kainic acid-induced degeneration of the neostriatum, Neurosci. Abstr., 4 (1978) 49.
32 21
22
23 24 25
Sanberg, P.R., Pisa, M. and Fibiger, H.C., Kainic acid injections in the striatum alter the cataleptic and locomotor effects of drugs influencing dopaminergic and cholinergic systems, Europ. J. Pharmacol., in press. Sanberg, P.R., Faulks, I.J., Anson, J.M. and Mark, R.F., Long term changes in emotional reactions and not in learning after intracerebral injections of cycloheximide or glutamate in neonatal chicks, Neurosci. Abstr., 6 (1980) 11. Sdraulig, R., Rogers, L.J. and Boura, A.L.A., Glutamate and specific perceptual input interact to cause retarded learning in chicks, Physiol. Behav., 24 (1980) 493 500. Wallnau, L.B. and Gallup, G.G., Jr., A serotonergic, midbrain-raphe model of tonic immobility, Biobehav. Rev.. 1 (1977) 35-43. Walsh, R.N. and Cummins, R.A., The open-field test: a critical review, Psychol. Bull., 83 (1976) 482-504.
19
Elsevier Biomedical Press
OPERANT, OPEN-FIELD AND TONIC IMMOBILITY BEHAVIOURS IN CHICKENS WITH FOREBRAIN INJECTIONS OF CYCLOHEXIMIDE OR G L U T A M A T E
P A U L R. S A N B E R G , I A N J. F A U L K S * , J U D I T H M. A N S O N and R I C H A R D F. M A R K
Department of Behavioural Biology, Research School of Biological Sciences, Australian National University, Canberra, ACT 2600 (Australia) (Received M a r c h 12th, 1981) (Accepted April 7th, 1981)
Key words: cycloheximide - glutamate - operant behavior - open-field - tonic immobility retarded learning - emotional reactions - fear - chicken
SUMMARY
Chickens that had received bilateral injections of cycloheximide or glutamate into the forebrain on day 2 of life and tested 4 weeks later showed no deficit in acquisition, performance or extinction of continuously reinforced appetitive key-pecking as compared to control birds injected with saline. However, chickens that had received injections of cycloheximide and were subsequently tested in an open-field apparatus took longer to leave the first square, defaecated more, and pecked, preened and moved about less than controls. They also showed longer durations of tonic immobility. Those injected with glutamate exhibited similar behaviour but were not significantly different from controls in the open field latency to leave the first square, defaecation, or tonic immobility tests. The above treatments have previously been described as producing permanent slowed learning in chickens on a pebble-floor task. Our results suggest that learning mechanisms may not be disrupted as shown by normal performance in a simple operant task but that enhanced emotionality or fear of novelty as revealed in the open field tests may interfere with the expression of learning behaviour in some situations.
* To w h o m all correspondence should be addressed. 0166-4328/82/0000-0000/$02.75 © Elsevier Biomedical Press
20 INTRODUCTION
In 1972 Rogers and Mark [14] reported that intracranial injections of cycloheximide (a cytoplasmic ribosomal protein synthesis inhibitor), given during the first week after hatching, would subsequently render the animal incapable of learning at the normal rate on a pebbled-floor task. Chicks injected with cycloheximide failed to reach the control levels of performance, that is they failed to choose predominantly grain, in preference to pebbles, in the latter half of the 60 pecks of the test. Sensory input (visual and auditory) within 3 h following the injections were required to slow subsequent acquisition of these behavioural tasks. Cycloheximide-treated birds were also slower to habituate to visual and auditory stimuli. In addition, Rogers and Anson [12] have shown that birds treated with cycloheximide persisted in pecking at preferred food more than saline-injected control animals. Rogers and her colleagues [12, 15] have found the injection parameters of dose, volume and age of chicks to be critical in obtaining all these effects. Hambley and Rogers [6] have postulated that cycloheximide produced these long-term behavioural changes in chicks by the decreased utilization for protein synthesis and consequent accumulation of the amino acids, glutamate and aspartate. Forebrain injections of these neurotransmitters, particularly glutamate, were shown to produce behavioural deficits on the pebble floor task qualitatively similar to those seen in the cycloheximide-injected birds [6, 23]. The factors that contribute to performance on the pebble floor are not clearly understood [10, 11]. Therefore, the present study was conducted in order to directly compare the long-term effects of neonatal injections of glutamate and cycloheximide on a variety of behavioural tasks conventionally used to measure changes in learning (appetitive key-pecking), activity (open-field) and emotionality (open-field and tonic immobility) in birds. A preliminary report has been presented elsewhere [22]. MATERIALS AND METHODS
(A) Material Birds and Housing White leghorn × Black Australorp male chickens were obtained from Research Poultry Farm (Victoria) on day 1 of life. They were housed in groups of 4-6 birds until day 14 when they were housed in pairs for the remainder of the study in 23 × 23 × 29 cm metal cages with a clear perspex front. Warmth (25 °C) and light were provided continuously by overhead lamps (25 W). Chicks had free access to grain (Hutmill, Victoria) and water (tap water with terramycin added) during the study, except when under deprivation schedules as described below.
21
Drug administration On day 2 of life, the chicks were randomly assigned to 3 groups and given bilateral intracerebral injections of either cycloheximide (20 /~g/hemisphere), glutamate (210 #g/hemisphere) or saline vehicle. The glutamate solution was buffered to pH 7.2 with Na2PO4. All injections were given in a vehicle of 25/~1 saline and performed freehand into the middle of each forebrain using a Hamilton 50/zl syringe. A 26.5-gauge needle was used and fitted with a rubber stopper to prevent penetration beyond 3 mm below the surface of the skull. These doses and volumes were used since they were found to provide the optimal conditions for producing pebble floor deficits [12].
Appetitive key-pecking Four standard BRS Foringer Skinner boxes for pigeons, enclosed in soundproof chambers, were used. Each was fitted with a centre illuminated pecking key located immediately above a grain feeder. They were modified with a grid floor of adjustable height, enlarged grain access hole and placement of a photocell beam across the food cubicle so that the beam was interrupted when the animal's head entered, as described elsewhere [16]. Programming and stimulus delivery were controlled by Digital K-logic solid state circuitry. Keypecks were recorded on printing counters converted from Royal 310P printing calculators. Photocell interruptions and reinforcements were recorded on counters converted from National Semiconductor 850A calculators [18]. Chicken starter crumble was used as the reinforcer.
Open-field A 80 x 100 cm platform with 50 cm high walls was used as the open-field. It was painted white and the floor was divided by black lines into 20 squares of 20 cm2. The open-field was placed in a 3.1 x 6.5 m laboratory with fluorescent lighting 2.8 m above the floor.
Tonic immobility Tonic immobility was measured on the floor of a BRS chamber with the door open for visual observation. (B) Methods
Acquisition and extinction of keyopecking At three weeks of age, 16 cycloheximide-, 10 glutamate- and 14 salineinjected chickens were restricted to 1.5 h of feeding daily sometime between 21.00 and 23.00 h. One week later the animals were given a 20 min magazine training session on both days 1 and 2. The animals were gently placed in the operant chamber, which had the grain feeder up and extra grain in the food
22 hopper (days 1 and 2) and scattered on a paper towel on the grid floor immediately before the hopper entrance (day 1). On days 3-5, the animals were given hopper training over 20 min sessions. On days 3 and 4 the feeder (hopper) delivered grain to the food cubicle for 10 sec followed by 5 sec of non-delivery. On day 5 food was delivered for 10 sec, followed by 20 sec of non-delivery. Photocell interruptions were measured over days 1-5. Days 6-10 consisted of autoshape-continuous reinforcement (CRF) training. The animals were given daily sessions of 30 min each, in which the hopper delivered grain for 5 sec every 30 sec (days 6 and 7) or 60 sec (days 8, 9 and 10). In each session the free grain delivery was preceded by 5 sec of red illumination of the pecking key, which ceased when the grain was delivered (autoshape). In addition, the chicken received 5 sec access to grain (CRF), if the key was pecked when not illuminated. A printing counter recorded those pecks which occurred during either the autoshape or CRF schedules over time (30 sec periods for days 6 and 7; 60 sec periods for days 8, 9 and 10). On days 11-19 the chickens were given CRF performance sessions only. These were 30 min long; the pecking key was illuminated red continuously, and only grain (5 sec access) contingent on key-pecking was available. On days 20-24 reinforcement was discontinued. Each of these 5 extinction sessions lasted until the chickens ceased to respond for 3 consecutive minutes. Both latency and number of responses to the extinction criterion were recorded.
Open-field At 4 weeks of age, 10 cycloheximide-, 10 glutamate- and 9 saline-injected chickens were placed individually in the central square of the open-field. For the next 5 min the following behaviours were observed: (1) the latency (sec) to leave the first square; (2) the number of squares entered by both feet (ambulation); (3) the number of pecks at the floor; (4) the number of preening episodes; and (5) the number of times the animals defaecated. The open field was cleaned and wiped with a weak vinegar solution between each test.
Tonic immobility At 4 weeks of age, 19 cycloheximide-, 17 glutamate- and 20 saline-injected chickens were tested individually for tonic immobility. The bird was held down by hand on its right side for 15 sec. Tonic immobility was measured as the time in sec from when the animal was released by the experimenter until it righted itself and rose to its feet. Additional induction attempts were given 60 sec after each unsuccessful attempt to elicit tonic immobility. If the animal failed to show tonic immobility over 5 attempts it was given a score of 0 sec. The floor was cleaned and wiped with a weak vinegar solution between each test. Over all behavioural experiments, treatment and testing were completely randomized.
23
Statistics One-way analyses of variance were used on the appetitive key-pecking acquisition data, open-field and tonic immobility data. Bonferoni [7] multiple comparison tests were used to test significance between groups. Analysis of variance for trends of trial means [2] were performed on the appetitive keypecking acquisition and extinction data. Because of skewness and heterogeneity of variance on magazine and hopper training, and open field latencies, nonparametric Mann-Whitney U-tests were used to test significance between groups. Two-tailed significance tests were used throughout the analyses. RESULTS
Appetitive key-pecking The median photocell counts for magazine and hopper training are shown in Fig. 1. Only on day 1 was there any significant effect. The birds treated with cycloheximide broke the photocell beam in the food cubicle less often than birds treated with saline (U = 61, nl/n2 = 14/16, P < 0.05). The performance of birds treated with glutamate and those given cycloheximide (U = 73, nl/n2 = 14/16, P < 0.05). The performance of birds treated with glutamate and those given cycloheximide 0d = 73, nl/n2 = 10/16, P >0.05) on this measure was not statistically different, although visual observation suggested that the birds MAGAZINE TRAINING DAY I
C
DAY 2
HOPPER TRAINING DAY 3
DAY 4
DAY 5
500
o U 400 I
U
•~ 300
e'Q,.
•
j
._~ 200 "(3
•
loo SAL CXM GLU
SAL CXM GLU
SAL CXM GLU
SAL CXMGLU
$AL CXMGLU
Fig. 1. Median responses of head entries in food cubicle during acquisition of magazine and hopper training. SAL, saline group; CXM, cycloheximide group; CLU, glutamate group.
24 9O
AUTOSHAPE
CRF
80 C
o
~
zo
u
60
@
~
o
//A
//A I //A //A //A //A //A //A //A
5O
\\~
4O
\\4 \
"5
\\Xl \\~ \\N
/ /A
\\'t
//A
\\N
30 //A zz A ///I
tO
/~A
20'
\-,
" "xl "4 \ \ ~ \\N
//11
//A //A
¢1
10
SAL
CXM
SAL
GLU
CXM
GLU
Fig. 2. Mean time and standard error of the mean to first keypeck response in the autoshape or continuous reinforcement (CRF) periods during acquisition of continuously reinforced key-pecking.
400
AUTOSHAPE-C R F
C RF
o--o SAUNE a--a CYCLOHEXIMIDE a--~ GLUTAMATE
350
c 0 ev
300 250
~.
200
~-
150
v O
O
100
I
I
I
I
I
I
i
I
[
I
1
I
I
I
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Meanresponsesduringacquisitionofcontinuouslyreinforced Days
Fig. 3. key-pecking. Autoshape-CRF means the session days in which autoshaping procedures were used.
25 treated with cycloheximide ate much less grain on day 1 than either of the other groups. The results of the autoshape-CRF are shown in Figs. 2 and 3. The mean latencies in total min to the first key-peek (Fig. 2) during the red illumination (autoshape) or non-illumination (CRF) periods were not significantly different between groups (F = 1.59, df = 3.37, P > 0.05, and F < 1, respectively). There were no differences between the mean response rates of the different groups (Fig. 3) (F < 1). All 3 groups showed a significant increase in the number of mean key-peck responses over the 13 days the birds were tested ( F = 52.4, d f = 13,481, P < 0.001). On any one day the mean key-peck response did not CRF E X T I N C T I O N 300['--I
cO t,. t,_
U
200-
I
SALINE CYCLOHEXIMIDE GLUTAMATE
o u G)
v
100-
Y/, Y/,
tO
0 4035 c
o
30
i,_ t,_
k.)
25
o 20 E tO
15 10
-I
5 0 2
3
4
5
Days
Fig. 4. Mean responses and standard error of the mean (upper panel) and mean latency (see) to criterion and standard error of the mean latency (lower panel) during extinction of continuously reinforced key-pecking.
26 differ significantly between the 3 groups (e.g. Day 19, F < 1). However, all 3 groups showed a significant decrease in key-pecking within each trial (e.g. Day 19, F = 15.79, df = 14,518, P < 0.001). Extinction of the key-pecking response is shown in Figs. 4 and 5. There were no significant differences between the mean time and the mean number of responses the groups took to reach the extinction criterion (Fig. 4). Analysis of variance of the key-pecking response revealed no significant effects of drug treatment (F = 1.42, df = 2,37, P > 0.05) or interaction of drug treatment with days ( F = 1.33, d f = 8,148, P >0.05). A significant effect over days, indicative of extinction behaviour, was noted ( F = 98.73, df = 4,148, P < 0.001). Analysis of variance of the latencies also revealed a non-significant effect of drug treatment (F = 1.06, df = 2,37, P >0.05) and a significant effect over session days ( F = 241.75, df = 4,148, P < 0.001). There were no significant differences
30
C R F EXTINCTION 25
20
o--o
o--a =--=
SALINE CYCLOHEXIMIDE GLUTAMATE
5 10
5
10
15
20
2.5
30
55
TriaJs
Fig. 5. M e a n responses over trials (1 trial = 60 sec) during extinction days 1 and 5 of continuously reinforced key-pecking.
27 between groups on each extinction day over time (Ps >0.05). The results of extinction sessions on days 1 and 5 with time are shown in Fig. 5.
Open-field Figure 6 shows that the birds treated with cycloheximide took significantly longer to leave the first square than birds treated with saline (U = 17, nl/n2 = 10/9, P <0.05), however, birds given glutamate did not differ from cycloheximide- or saline-treated animals (U = 48, nl/n: = 10/10, and U = 28, n~/n2 = 10/9 respectively, Ps >0.05). There was a significant effect of groups on the ambulation measure ( F = 5.3, df = 2,26, P < 0.02). Multiple comparison tests revealed that both the cycloheximide- ( F = 9 . 1 4 , d f = 1,26, P <0.02) and glutamate-treated ( F = 6.91, d f = 1,26, P < 0.05) groups passed significantly less squares than did controls. There was no significant difference between cycloheximide or glutamate scores (F < 1) (Fig. 6). Additional open-field data is shown in Fig. 7. There were significant group effects for the other behavioural measures (Fig. 7). Multiple comparison tests revealed that both the cycloheximide-treated and glutamate-treated groups pecked at the floor (Fs = 20.49 and 10.15, df = 1,26, Ps < 0.0005 and 0.02, respectively) and preened (Fs = 7.32
280
LATENCY TO LEAVE FIRST SQUARE
I
260
v~ 240
i
- 0¢..
o
~
220
¢,.
200
180
160
AMBULATION
16
O 12
Z c
3 o 8
.Q cQ
4
Q
2
140
o
0 SAL
CXM
GLU
SAI
CXM
GLU
Fig. 6. Latency to leave first square and ambulation in saline (SAL)-, cycloheximide (CXM)- and glutamate (GLU)-injected chicks tested for 5 min in an open-field. Bars represent standard errors of the means.
28
PECKS AT GROUND
4.0
PREENING -EPISODES
DEFAECATION
3.5
e-
3.0
> LU
,_,_ o
2.5
..a E
2.0
Z e.-
1.5
0
~
1.o 0.5
SAL
CXM
GLU
SAL
CXM
GLU
SAL
CXM
GLU
Fig. 7. Number of pecks at ground, preening episodes and faecal boli in saline (SAL)-, cycloheximide (CXM)- and glutamate (GLU)-injected chicks tested for 5 min in an open-field. Bars represent standard errors of the means.
and 10.81, d f = 1,26, Ps < 0.05 and 0.01, respectively) less than saline-treated birds. On the other hand, birds treated with cycloheximide and glutamate tended to defaecate more than saline-treated birds although the scores did not quite reach significance (Fs =4.48 and 6.06, df--1,26, Ps <0.13 and 0.06, respectively). There were no differences between birds treated with cycloheximide and glutamate in these three behavioural measures.
Tonic immobility Analysis of variance of the tonic immobility results (Fig. 8) showed a significant group effect ( F = 3.42, df--2,53, P < 0.05). Multiple comparison tests revealed a significant difference between cycloheximide- and saline-injected birds ( F = 6.57, df = 1,53, P < 0.05). Birds injected with glutamate were not significantly different from either cycloheximide- (F = 2.93, df = 1,53, P > 0.05) or saline-treated birds (F < 1). DISCUSSION
Neonatal injections ofcycloheximide or glutamate did not alter either rate of acquisition or asymptotic performance of key-pecking behaviour in chickens.
29 TONIC IMMOBILITY 800 700 600
500 U 4)
,- 400 3O0 200 100
m
SAL
CXM
GLU
Fig. 8. Duration of tonic immobility in saline (SAL)-, cycloheximide (CXM)- and glutamate (GLU)-injected chicks. Bars represent standard errors of the means.
Regardless of treatment, all groups performed similarly throughout hopper, autoshaping and CRF training procedures. Furthermore, there were no differences between groups on extinction of the CRF schedule. Only on the first day of magazine training did the group treated with cycloheximide respond less than the other groups. This difference probably reflects an enhanced fear response to novel situations, since chickens treated with cycloheximide show enhanced tonic immobility. Tonic immobility is a reliable measure of emotionality and fear in chickens [3, 4] and has been shown to be negatively correlated with depressed magazine training [19]. Tonic immobility is also inversely proportional to the position of each bird in the peck-order [3]. Chickens treated with cycloheximide fall to the bottom of the peck-order [15] and display altered levels of emotionality when reacting to visual and auditory stimuli (Anson, unpublished data). Anson (unpublished data) also found a pattern of responding consistent with enhanced emotionality in cycloheximide treated birds tested on the pebble floor task; that is, they took significantly longer to make the first 5 pecks than birds treated with saline. Birds treated with cycloheximide or glutamate were less active in an open-field situation than controls as reflected by fewer ambulatory, pecking and preening episodes. This relative immobility may indicate increased fear and general emotionality, since both groups tended to defaecate more than salineinjected birds. Some investigators have questioned the use of defaecation as a
30 measure of emotionality [1, 25], while the latency to leave the first square is considered a better measure [9]. Birds treated with cycloheximide, but not glutamate, showed significantly longer latencies to leave the first square than did controls. A similar pattern of results was also found in the tonic immobility experiment. The differentiation between the cycloheximide and glutamate effects on the two emotionality measures of open-field latency and tonic immobility, compared to those induced by saline, could mean that the long-term effects of these drugs are produced by different mechanisms. The fact that the differences between glutamate and cycloheximide over all behavioural tests were not significant may imply that the effects of glutamate injections alone are simply not as strong as cycloheximide-induced effects. Cycloheximide injections increase the brain levels of aspartate and other amino acids in addition to that of glutamate [6]. Previous studies with the pebble floor task [6, 12-15, 23] have shown that birds injected with cycloheximide or glutamate require many more pecks than saline-injected chicks to reach a similar level of performance. These results were interpreted as indicating that cycloheximide and glutamate interfered with associative and/or memory processes to produce slowed learning. In the present study, treatment with cycloheximide or glutamate had no significant effect on the rate of learning of an appetitive instrumental response. Perhaps alterations in pebble floor performance after treatment with cycloheximide or glutamate reflect not a learning deficit but altered emotionality. The results from the open-field study and the tonic immobility experiment suggest this could be the case. However, changes in the general emotional state of an animal alone cannot easily explain why [13, 23] perceptual input for the 3 h post-injection period is necessary for the drugs to alter subsequent behaviour on the pebble floor learning task. More work is needed before firm conclusions can be made about what effects cycloheximide and glutamate are having on the brains and behaviour of young chickens. Originally, Rogers et al. [15] reported that cycloheximide injections did not produce light or electron microscopic evidence of cellular damage in the hippocampal and Wulst areas. However, since glutamate is a known neurotoxic agent, especially in neonatal animals [5, 8] more detailed studies in other areas need to be carried out to elucidate the potential histological effects of cycloheximide and glutamate forebrain injections in young chickens. Previous research has shown permanent changes in neurotransmitter pharmacology following injection of kainate, an analogue of glutamate, into the rat striatum [17, 21] resulting in permanent changes in activity and emotionality [20]. The known role of monoaminergic systems in emotionality [24] suggest that studies investigating permanent changes in this transmitter system following cycloheximide or glutamate injections may be fruitful.
31 ACKNOWLEDGEMENTS
The authors appreciate the excellent assistance of Drs. K. Gillette, J. Irwin, W.P. BeUingham, I.G. Morgan and K.E. Sanberg throughout the course of this study. This study was completed in partial fulfilment of the Doctor of Philosophy degree from the Australian National University to P.R. Sanberg. REFERENCES 1 Archer, J., Tests for emotionality in rats and mice: a review, Anita. Behav., 21 (1973) 205-235. 2 Edwards, A.L., Experimental Design in Psychological Research, Holt, Rhinehart and Winston, New York, 1972, pp. 334-338. 3 Gallup, G.G., Jr., Tonic immobility as a measure of fear in domestic fowl, Anim. Behav., 27 (1979). 4 Gallup, G.G., Jr., Ledbetter, D.H. and Maser, J.D., Strain differences among chickens in tonic immobility: evidence for an emotionality component, J. comp. physiol. Psychol., 90 (1976) 1075-1081. 5 Garattini, S., Evaluation of the neurotoxin effects of glutamic acid. In R.J. Wurtman and J.J. Wurtman (Eds.), Nutrition and Brain, Vol. 4, Raven Press, New York, 1979, pp. 79-124. 6 Hambley, J.W. and Rogers, L.J., Retarded learning induced by intracerebral administration of amino acids in the neonatal chick, Neuroscience, 4 (1979) 677-684. 7 Miller, R.G., Simultaneous Statistical Influence, McGraw-Hill, New York, 1966. 8 Olney, J.W., Brain damage and oral intake of certain amino acids, Advanc. exp. Med. Biol., 69 (1976) 497-506. 9 Ossenkopp, K.-P., The open-field test as a rapid and sensitive behavioural measure in white peking ducklings Anas platyrhynchos, Bird Behav., 2 (1980) 23-35. 10 Reymond, E. and Rogers, L.J., Deprivation of the visual and tactile aspects of food important to learning performance of an appetitive task by chicks, Behav. Neural Biol., 31 (1981) 425-434. 11 Reymond, E. and Rogers, L.J., Diurnal variation in learning performance of young chicks, Anim. Behav., 29 (1981) 241-248. 12 Rogers, L.J. and Anson, J.M., Cycloheximide produces attentional persistence and slowed learning in chickens, Pharmacol. Biochem. Behav., 9 (1978) 735-740. 13 Rogers, L.J. and Drennen, H.D., Cycloheximide interacts with visual input to produce permanent slowing of visual learning in chickens, Brain Res., 158 (1978) 479-482. 14 Rogers, L.J. and Mark, R.F., Retarded learning in chicks after brief inhibition of brain protein synthesis, Proc. Aust. Physiol. Pharmacol. Soc., 3 (1972) 119. 15 Rogers, L.J., Drennen, H.D. and Mark, R.F., Inhibition of memory formation in the imprinting period: irreversible action of cycloheximide in young chickens, Brain Res., 79 (1974) 213-233. 16 Sanberg, P.R., Measuring feeding responses in operant research, Psychol. Rep., 45 (1979) 942. 17 Sanberg, P.R., Haloperidol-induced catalepsy is mediated by post-synaptic dopamine receptors, Nature (Lond.), 284 (1980) 472-473. 18 Sanberg, P.R. and Bellingham, W.P., Digital counters: inexpensive alternatives, Physiol. Behav., 23 (1979) 795-797. 19 Sanberg, P.R., Faulks, I.J., Bellingham, W.P. and Mark, R.F., Relationship between tonic immobility and operant conditioning in chickens Gallus gallus, Bird Behav., 3 (1981) 51-56. 20 Sanberg, P.R., Pisa, M. and Fibiger, H.C., Locomotor activity, exploration and neophobia in rats with kainic acid-induced degeneration of the neostriatum, Neurosci. Abstr., 4 (1978) 49.
32 21
22
23 24 25
Sanberg, P.R., Pisa, M. and Fibiger, H.C., Kainic acid injections in the striatum alter the cataleptic and locomotor effects of drugs influencing dopaminergic and cholinergic systems, Europ. J. Pharmacol., in press. Sanberg, P.R., Faulks, I.J., Anson, J.M. and Mark, R.F., Long term changes in emotional reactions and not in learning after intracerebral injections of cycloheximide or glutamate in neonatal chicks, Neurosci. Abstr., 6 (1980) 11. Sdraulig, R., Rogers, L.J. and Boura, A.L.A., Glutamate and specific perceptual input interact to cause retarded learning in chicks, Physiol. Behav., 24 (1980) 493 500. Wallnau, L.B. and Gallup, G.G., Jr., A serotonergic, midbrain-raphe model of tonic immobility, Biobehav. Rev.. 1 (1977) 35-43. Walsh, R.N. and Cummins, R.A., The open-field test: a critical review, Psychol. Bull., 83 (1976) 482-504.