- Email: [email protected]
Visual input and lateralization of brain function in learning in the chick
Neuroscience Vol. 52, No. 2, pp. 393--401,1993 Printed in Great Britain
0306-4522/93$6.00+ 0.00 PergamonPress Ltd © 1992IBRO
VISUAL INPUT A N D LATERALIZATION OF BRAIN FUNCTION IN LEARNING IN THE CHICK C. SANDI,*T. A. PATTERSONand S. P. R. ROSE Brain and Behaviour Research Group, Department of Biology, The Open University, Milton Keynes MK7 6AA, U.K. Abstract--Several lines of evidence (biochemical,neuroanatomical, electrophysiological,and behavioural) have indicated a critical role for the intermediate medial hyperstriatum ventrale of the chick forebrain in the acquisition of a passive avoidance response. Previous lesion studies indicated that bilateral or left, but not right, pretraining intermediate medial hyperstriatum ventrale lesions interfere with the acquisition of this task. We have further analysed this asymmetrical involvement of the intermediate medial hyperstriatum ventrale by use of a monocular learning protocol and intermediate medial hyperstriatum ventrale lesions (sham, bilateral, or unilateral). The results indicated that there is interocular transfer of information of passive avoidance learning between the two eye systems, with a tendency to be more successful from the fight eye system to the left than in the opposite direction. As in binocular conditions, bilateral pretraining intermediate medial hyperstriatum ventrale lesions impair learning in monocularly trained animals. Unilateral lesions to either left or right monocularly trained experimental animals resulted in amnesia when they were made to the right intermediate medial hyperstriatum ventrale and the chicks were trained/tested with the left eye open. These results indicate that, although right intermediate medial hyperstriatum ventrale lesions do not result in amnesia in binocular animals, this region is capable of participating in memory acquisition processes. They also suggest a connection between lateralization of intermediate medial hyperstriatum ventrale function in passive avoidance learning and the behavioural and structural visual asymmetries known to occur in chicks.
The demonstration of functional hemispheric asymmetries in humans43 has stimulated research dealing with cerebral lateralization both in human and nonhuman species. 22 Although the phenomenon was initially considered specific to humans, during the past two decades a number of behavioural and biological asymmetries have been reported in other species) 8 Some of the functional asymmetries described in laboratory mammals 1°,~9are characterized by a marked stability over time for a specific individual but a varied distribution in direction and/or magnitude across the population. By contrast, most of the functional asymmetries observed in avian species have been found to be directionally consistent in the p o p u l a t i o n y °,27,37 resembling in this respect the well-established asymmetries of handedness and language in humans. In particular, the avian visual system provides a suitable tool for studying information processing in the separate brain hemispheres since the optic nerves virtually show a complete decussation at the chiasmatic level) 3 Using the domestic chick to study the neural mechanisms involved in early types of learning (e.g. imprinting, one-trial passive avoidance learning), a variety of asymmetric biochemical, morphological, *To whom correspondence should be addressed. Abbreviations: IMHV, intermediate medial hyperstriatum
ventrale; MeA, methylanthranilate.
and electrophysiological changes23,25,4°,~ have been found in different brain areas, and especially the intermediate medial hyperstriatum ventrale (IMHV), both in untrained and trained birds. In particular, following one trial passive avoidance training (in which chicks, which peck spontaneously at small beads, learn after a single trial to avoid a bead coated with a bitter substance), enhanced metabolic activity4~ and a consequent biochemical cascade have been shown to occur asymmetrically in the left IMHV. Ls,4~ Several measures of dendritic and synaptic morphology and number are specifically enhanced in the same region. 2s'44'45The strongest evidence for a lateralized involvement of the IMHV in learning stems from psychopharmacological and lesion studies. Thus, injection of amnestic agents, such as inhibitors of protein synthesis29 or protein kinase C 8 into the left, but not the right, IMHV results in amnesia for the passive avoidance. Left, but not right, pretraining IMHV lesions result in amnesia in chicks tested 3 h after training.3° However, post-training bilateral IMHV lesions do not interfere with retention of the task (though they do prevent discrimination of the bead based on colour), 31 suggesting that a memory trace once formed in the left IMHV is subsequently dispersed to other brain regions. However, not all changes are confined to the left IMHV; in the minutes to hours after training a variety of changes also occur bilaterally or in the right IMHV. 3,24,42
393
C. SANDI et al.
394
The aim of the present study was to use a m o n ocular learning protocol to analyse this asymmetrical involvement of the I M H V in the acquisition o f passive avoidance learning. Chicks have two m a j o r visual pathways, one projecting from the optic tectum to the ectostriatum a n d a n o t h e r projecting in a n asymmetric m a n n e r from the t h a l a m u s to the visual Wulst or hyperstriatum. 7,2~'39Visual information a b o u t the training experience is one of the m a j o r c o m p o n e n t s of passive avoidance learning, since trained chicks are able to discriminate between similar beads differing in some of their visual characteristics (for example, colour) 3t a n d the I M H V is considered to be a n area critically involved in certain kinds of visual recognition. 14 T h e present experiments were designed to study: (i) the role of the I M H V in interocular transfer of i n f o r m a t i o n of passive avoidance learning between the two eye/brain-related systems; a n d (ii) the possibility t h a t the right I M H V could participate in learning if visual i n f o r m a t i o n is forced to arrive at the right hemisphere by occluding the right eye.
EXPERIMENTAL PROCEDURES
were placed in pairs into pens (20 x 25 cm) with scattered chick crumbs. The pens were maintained at 28-30°C and were illuminated by an overhead red light (25 W). After the period of equilibration, the chicks were pretrained by three 10-s presentations of a small (2.5 mm) white bead, and then trained by a single presentation of a 4-mm bright chrome bead coated with methylanthranilate (MeA). Only chicks that pecked in at least two of the pretraining trials and, during training, evinced a clear disgust response (head shaking, bill wiping, etc.) were considered as experimental subjects (about 80% of chicks). When the experiments required changing the eye-patch from one eye to the other (Experiments I, 4), this was done 1 h after training. In all experiments, 3 h after training chicks were tested by being offered a dry chrome bead for 30 s, and their response (peck or avoid) was noted. To accumulate enough data to evaluate the effects of the treatments, several replications of each experiment (five to nine replications per experiment carried out throughout a nine-month period), including a balanced number of animals in each group, were made. The behavioural results were compared using the chi-squared test of independence. After training and testing in each experiment involving lesions, the chicks were killed and their brains were examined for sites of lesions using normal histological techniques. Both testing and histological examination were conducted by an experimenter blind to the past history of the chicks. Results from chicks with incorrectly placed lesions were excluded.
Animals Ross Chunky chicks (Gallus domesticus) of both sexes were hatched from commercially obtained eggs in communal brooders and maintained on a 12:12 h light
Surgical procedures Anaesthesia was induced by intraperitoneal injection of Equithesin (0.28 ml/100 g body weight). When narcosis became apparent, the chick was transferred to a stereotaxic instrument where its temperature was maintained by an electric underblanket. When the bird was fully anaesthetized, its cranium was secured in a modified small animal headholder, with the beak bar locked at 5 mm below and 11.5 mm anterior to the central axis of the ear bar. This orientation is similar to that used in Youngren and Phillip's47 atlas of the three-day-old chick brain. The scalp and underlying muscle insertions were deflected and small craniotomies were made in the skull directly over the electrode placement sites. After incision of the dura, the temperature-sensitive electrode was lowered under stereotaxic control. The coordinates used for the IMHV were 0.4 mm anterior to the ear bars, 0.9 mm lateral to the midline, and 1.7 mm ventral to the surface of the brain. Lesions were induced by radiofrequency using a Radionics radio frequency lesion generator (Model RFG4; Radionics, Inc., Burlington, MA). The temperature at the electrode tip was maintained at 60°C for 90 s. The electrode was removed when the temperature of the electrode probe returned to body temperature. The skull flap was closed and secured with warm bone wax. The scalp was sutured, and the chick was left to recover in a warm box for several hours. The chicks were then given water and allowed to recover in these boxes for several hours before training or testing. Sham control chicks underwent identical procedures to those of the lesioned animals, however, no current was passed through the electrode tip.
Training procedures The chicks were 24 h old at training. One hour before training, either the left (L) or the right (R) eye (cf. specific protocols) was occluded by means of a black paper patch, lightly stuck on to the down around the eye. Then, the chicks
RESULTS
Histology A histological reconstruction of a typical bilateral I M H V lesion is s h o w n in Fig. 1. Unilateral lesions were of the same size and at the same position o n the two sides of the brain as the bilateral lesions. The location and extent o f the lesions were similar a m o n g the different experimental groups as well as equivalent to those o b t a i n e d in previous studies involving lesions to this a r e a ) 7'3°
Experiment 1 Monocular learning, and testing the experienced or naive eye. A first experiment was carried o u t in order to establish the level of interocular transfer of the passive avoidance task u n d e r the m o n o c u l a r conditions chosen for these experiments. Twentyf o u r - h o u r old chicks were removed from the b r o o d e r s a n d either the left or right eye was covered with a n eye-patch. One h o u r after training, the eye-patch was carefully removed a n d a new one placed covering either the same eye again (in h a l f of the chicks in each m o n o c u l a r condition) or the opposite eye (in the other half o f the chicks in each m o n o c u l a r condition) (see Fig. 2). Since different sources of stress a r o u n d the training trial are k n o w n to interfere with the acquisition o f the task (Patterson, u n p u b l i s h e d observations), the schedule for allocating a n d c h a n g i n g the eye-patches (1 h prior a n d 1 h after the training trial, respectively) was selected to allow the chicks to h a b i t u a t e to the eye-patches a n d to reduce possible stress a r o u n d the training session. Pilot studies indicated t h a t eye-patches did not influence pecking in animals trained o n a water-coated bead.
IMHV lesions and functional lateralization HA
395 l-IV
A 3.5 HA
HV I-IV
A 6.0
@
HA
HV
A 3.0 HV
2.5 A 4.5 .V
A 4.0 Fig. I. Typical extent of bilateral IMHV lesions included in this study. Solid areas indicate total lysis of tissue, shaded areas indicate scar tissue degenerating neuropil. Unilateral lesions were of similar size and at the same position on the two sides of the brain as the bilateral lesions. A, archistriatum; E, ectostriatum; HA, hyperstriatum accesorium; HV, hyperstriatum ventrale; LPO, lobus parolfactorious; N, neostriatum; V, ventricle.
There was no difference between the two groups trained and tested with the same eye (~2 = 0.76, n.s.); i.e. both groups showed good levels of retention (left eye-patch t r a i n i n g + left eye-patch testing, LL = 82.1% of avoidance responses, n = 28; R R = 69.0%, n = 29). However, in the conditions involving testing with the untrained eye, the group trained with the left eye and tested with the right eye. (RL = 56.4%, n = 39) showed reduced avoidance as compared with the group trained with the right eye and tested with the left eye (LR = 77.5%, n = 40) (X2= 3.98, P = 0.05). Thus, transfer of information from the left
eye system to the right is slightly poorer than in the opposite direction (see Fig. 2).
Experiment 2 Effects of bilateral intermediate medial hyperstriatum ventrale lesions on monocular learning. Early monocular deprivation has been reported to induce different types of neural plasticity in the avian fore° brain. 4,n Therefore, before studying the effects of unilateral I M H V lesions (Experiments 3 and 4), we determined if the I M H V is necessary for the monocular acquisition of passive avoidance learning, as has
C. SANDl et al.
396 TRAIN
(~[~) ~
TEST
PREDICTIONS
RESULTS
n
% Avoidance
14 13 11 11
78.6 76.9 36.4* 45.4*
~D
, Recall
Recall
Sham lesion + left eye-patch Sham lesion + right eye-patch Bilat IMHV lesion + left eye-patch Bilat IMHV lesion + fight eye-patch
(~ [~o
,Recall
Recall
*P < 0.05 vs corresponding sham-lesioned group in the same eye-patch condition
I~[~
JRecall
Reduced Recell
Fig. 2. Protocol, predictions and results of Experiment 1 designed to establish the level of interocular transfer of the passive avoidance response when chicks are trained with one eye viewing, either left or right, and tested with the opposite eye. Open ellipse, eye open; filled ellipse, eye occluded; open circle, sham IMHV lesions; filled circle, genuine IMHV lesion. been shown to be the case in binocular experimental animals.~5,3° This was done by testing whether bilateral ablation of the I M H V prior to training interfered with the acquisition of passive avoidance learning in monocularly trained and tested chicks. At 18 h old, the chicks were given sham or bilateral I M H V lesions. One hour before training, chicks in each condition were divided into two experimental groups with either left or right eye covered, trained and tested 3 h later. The eye-patch was maintained in the same place until the end of the experiment (see Fig. 3). The results of this experiment are shown in Table 1. Just as the unoperated birds in Experiment 1, sham-operated animals displayed similar levels of avoidance with either eye occluded (n = 18 chicks per group, 3(2= 0.0, n.s.). However, pretraining bilateral I M H V lesions impaired the acquisition of passive avoidance learning in monocularly, left (n = 15, Z2=4.90, P <0.005) or right (n = 2 0 , Z2=5.39, TRAIN
Table 1. Effects of sham or bilateral intermediate medial hyperstriatum ventrale lesions on passive avoidance learning in monocular, left or right trained and tested chicks
TEST
PREDICTIONS
P < 0.025), trained and tested animals. These results indicate that, as is the case in the binocular conditions, ~5'3° an intact I M H V is necessary for the successful acquisition of passive avoidance learning using only one eye.
Experiment 3 Effects o f unilateral intermediate medial hyperstriatum ventrale lesions on monocular learning using the same eye at training and at testing. Having established that bilateral lesions to the I M H V result in amnesia for monocular learning, the next question was whether in a monocular learning situation there would be a lateralized involvement of the I M H V depending on the eye used to perform the task. Chicks were given unilateral, left or fight, lesions and trained either with the left or with the right eye open. Because left but not right I M H V lesions result in disruption of learning in birds viewing the bead binocularly, 3° we made the following predictions (see Fig. 4): (i) left I M H V lesions given to chicks trained with the fight eye open (left eye-patch) should render the chicks amnesic for the task; (ii) lesions performed in the I M H V ipsilateral to the open eye, either left or right (and so contralateral to the eye-patch), would not have any effect on learning, unless the left I M H V is critical for the associative processing related to the different components (i.e. visual, gustatory, emotional) of the task, in which case left I M H V lesions could
RESULTS TRAIN
TEST
PREDICTIONS
RESULTS
(Amnesia?)
(~)
, Q~[~)
, Amnesia
Amnesic
Fig. 3. Protocol, predictions and results of Experiment 2 designed to determine whether the IMHV is necessary for the monocular acquisition of passive avoidance learning. Lesions were made pretraining on the day the chicks hatched, the chicks were trained the following day and tested 3 h subsequently. Open ellipse, eye open; filled ellipse, eye occluded; open circle, sham IMHV lesions; filled circle, genuine IMHV lesion.
Fig. 4. Protocol, predictions and results of Experiment 3 designed to determine whether there is a lateralized involvement of the IMHV on passive avoidance learning depending on the eye used to perform the task. Open ellipse, eye open; filled ellipse, eye occluded; open circle, sham IMHV lesions; filled circle, genuine IMHV lesion.
IMHV lesions and functional lateralization Table 2. Effects of unilateral left or right intermediate medial hyperstriatum ventrale lesions in chicks trained/ tested with one eye, left or right, occluded Left IMHV lesion + left eye-patch Left IMHV lesion + right eye-patch Right IMHV lesion + left eye-patch Right IMHV lesion + right eye-patch
n
% Avoidance
23 19 19 19
65.2 78.9 78.9 42.1"
*P < 0.05 vs left IMHV-lesioned + right eye-patch and vs right IMHV lesioned + left eye-patch groups. also impair learning in left eye-trained/tested chicks; and (iii) if the right IMHV is able to take over the role of the left IMHV in learning acquisition processes, when the visual information is made to arrive mainly to the right hemisphere by occluding the right eye, right IMHV lesions would result in amnesia in chicks trained/tested with the left eye (right eye-patch); but if this take-over is not possible, the chicks would perform successfully in the task. At 18 h old, the chicks were given one of two types of lesion: either (i) a lesion in the left IMHV and a sham lesion in the right IMHV (left IMHV lesion group); or (ii) a lesion in the right IMHV and a sham lesion in the left IMHV (right IMHV lesion group). All other procedures were similar to those of Experiment 2. The results of this Experiment are shown in Table 2. When chicks were given left IMHV lesions, there were no significant differences between the right eye- (n = 23) and left eye- (n = 19) trained groups (Z2= 0.41, n.s.); i.e. both groups showed high levels of avoidance at retention. However, when the lesions were made to the right IMHV, chicks trained with the left eye (n = 19) displayed significantly lower levels of avoidance than chicks trained with the right eye (n = 19) which displayed a good retention level (~2 = 3.96, P < 0.05). In addition, the right IMHVlesioned group in the left-eye open condition also differed significantly from the left IMHV-lesioned group in the same eye viewing condition (X2 = 3.96, P < 0.05), but it was no different from the left IMHV lesion plus right-eye open group (X2= 2.24). There were no differences between the performance of the right IMHV lesion and left eye-patch group and left IMHV-lesioned animals in either monocular training/ testing condition, right-eye open (Z2= 0.41, n.s.) or left-eye open (~(2= 0.00, n.s.) (see Fig. 4). Therefore, our first prediction about an amnesic effect for left IMHV lesions given to chicks performing with the right eye seeing (left eye-patch) was not supported by the present findings. There was, however, supporting evidence for the second prediction in that lesions in the IMHV ipsilateral to the eye open, regardless of which eye it is, would not have any effect on learning. The fact that performance of chicks with left IMHV lesions and performing with the left eye open (right eye-patch) (main visual information arriving to the right hemisphere) was not impaired seems to indicate that the right IMHV could be critically involved in the acquisition processes of the
397
task when the information is forced to be, at least primarily, processed in the right hemisphere. The strongest evidence for such a role comes from the findings related to the third prediction, showing that when information is forced to arrive at the fight hemisphere (right eye-patch group), right IMHV lesions result in amnesia. In summary, these striking results indicated that in our monocular learning protocol, left IMHV lesions did not interfere with learning regardless of the eye used, but fight IMHV lesions impaired performance in the left eye open group. Thus, what it is revealed is a lateralized effect of IMHV lesions opposite to that found in binocular learning) ° Could this result be a consequence of the known asymmetry in the visual pathways (see Discussion)? To test this hypothesis, we reasoned that if unilateral, left or right, IMHVlesioned chicks were monocularly trained under the same conditions as in Experiment 3 but were tested for retention using the untrained eye, the results should reflect this asymmetry in the visual projections (Experiment 4).
Experiment 4 Effects of unilateral intermediate medial hyperstriatum ventrale lesions on monocular learning using a different eye at training and at testing. Chicks given unilateral IMHV lesions were monocularly trained and then tested using the opposite eye. Based on the results obtained in the previous experiments the predictions were (see Fig. 5): (i) left IMHV lesions would not interfere with performance of chicks trained with the right eye and tested with the left eye (left eyepatch training + right eye-patch testing); (ii) chicks given left IMHV lesions and performing at training with the left eye and at testing with the right eye (right eye-patch training + left eye-patch testing), may show impairment in retention since interocular transfer from the left to the right eye was already shown to be slightly poorer in non-lesioned chicks in Experiment 1; (iii) if the lesions are given to the right IMHV and the chicks are trained with the right eye, in accordance with Experiment 3, the chicks should successfully acquire the task; however, if they are tested for retention with the left eye (left eye-patch training + right eye-patch testing) they should be now amnesic since the left eye is not well connected with the left hemisphere (see Discussion; also see Refs 7, 39); and (iv) since chicks given right IMHV lesions and trained with the left eye (which feeds to the right hemisphere) were shown to be amnesic (Experiment 3), if they are now tested with the right eye (right eye-patch training + left eye-patch testing), they should also be amnesic since they would have not learnt the task. Chicks were given unilateral IMHV lesions as in Experiment 3. As before, birds in each lesion condition were divided into experimental groups with either left or right eye occluded 1 h before training. There were therefore four groups including all the combinations between lesion x eye-patch conditions. By contrast to
C. SANDI et al.
398 TRAIN
TEST
,
PREDICTIONS
o
Recall
RESULTS
Recall
IMHV lesions, in chicks performing monocularly in a passive avoidance task, are asymmetric (see Fig. 5). They also confirm the result of Experiment 1 and Gaston ~6 about a mild detrimental effect for interocular transfer when the information learned with the left eye has to be retrieved with the fight eye. DISCUSSION
The left IMHV has been regarded as a structure necessary for successful acquisition of passive avoidance learning. On the basis of a wide body of data showing consistent asymmetrical learning-induced Fig. 5. Protocol, predictions and results of Experiment 4 biochemical and cellular changes in the IMHV in designed to determine whether there is an asymmetrical effect of the involvement of the IMHV on retention for chicks exposed to a passive avoidance task (see passive avoidance learning depending on the eye used at Introduction), and our previous finding that left, but training and at testing. Open ellipse, eye open; filled ellipse, not fight, IMHV lesions interfere with acquisition of eye occluded; open circle, sham IMHV lesions; filled circle, the avoidance response) ° Although a set of subsegenuine IMHV lesion. quent lesion experiments suggested a possible role of the right IMHV in distributing the information Experiment 3, 1 h after training the eye-patches were acquired from the left IMHV to other brain areas, ~7 transferred to the opposite eye. The results of this the physiological basis for such a lateralization of experiment are shown in Table 3. As can be seen, IMHV function in learning remained obscure. It chicks receiving left IMHV lesions and performing in could be that the left IMHV itself is uniquely capable the right-eye open at training + left-eye open at test- of learning-induced plasticity, or, alternatively, that ing conditions (n = 19) showed good levels of avoid- it assumed this role as a consequence of prior aspects ance. Left IMHV-lesioned chicks in the left-eye open of hemispheric asymmetry such as, for example, at training + right-eye open at testing group (n = 13) visual processing. Under our experimental monocular conditions, showed a slight but non-significant trend to reduced avoidance scores by comparison with their IMHV- there was interocular transfer of information of passive lesioned counterparts in the opposite eye viewing avoidance learning, although this transfer tended to condition (Z2= 1.46, P < 0.25). When the lesions be less successful from the left eye to the right eye were given to the right IMHV, both groups showed than in the opposite direction (Experiments 1 and 4). low levels of avoidance as compared with the "left Given that bilateral IMHV lesions were shown to interfere with learning in monocularly trained chicks IMHV lesion and right-eye open at training + left-eye open at testing" group showing normal retention (Experiment 2), the lack of effect of left IMHV lesions observed under monocular conditions (Experiment 3) levels (right IMHV lesion, left eye-patch training + fight eye-patch testing: n = 21, Z2 =4.99, P < 0.03; could be interpreted as the consequence of an involveright IMHV lesion, right eye-patch training + left ment of the fight IMHV in learning processes. The eye-patch testing: n = 19, X2= 3.84, P = 0.05), but fact that only fight IMHV lesions disrupted learning there was no difference between the two monocular when chicks performed with the left eye, but left IMHV lesions were ineffective even when chicks conditions analysed (X2= 0.05, n.s.). Therefore, these results confirm the predictions performed with the contralateral, right eye suggests made for each experimental condition on the basis of a possible relation between the lateralised effects the hypothesis that the laterality effects of unilateral of IMHV lesions and the asymmetry of visual projections. Stronger support for this interpretation was Table 3. Effects of unilateral left or right intermediate obtained when chicks trained with the fight eye, which medial hyperstriatum ventrale lesionson avoidance scores in are known to acquire the task successfully either with chicks trained with one eye and tested with the opposite one a left or a right IMHV lesion (Experiment 3), were n % Avoidance tested for retention with the left eye (Experiment 4). Under such circumstances, chicks with a lesion in the Left IMHV lesion + left eye-patch 19 73.7 training + right eye-patch testing fight, but not in the left, IMHV showed amnesia for the task. Left IMHV lesion + right eye-patch 13 46.2 It is a peculiar feature of the chick, during the first training + left eye-patch testing three weeks of life, that there is an asymmetry in the Right IMHV lesion + left eye-patch 21 33.3* visual thalamofugal projections; that is, the fight eye training + right eye-patch testing projects to both left and fight hemispheres, whereas Right IMHV lesion + right eye-patch 19 36.8* the left eye is connected almost only with the fight training + left eye-patch testing *P < 0.05 vs corresponding eye-patch group in the left hemisphere. 7,39 This asymmetry suggests that one explanation for our results is that when the right eye IMHV lesion condition.
IMHV lesions and functional lateralization is monocularly trained, information about the task will be accessible for processing in both hemispheres but that the ipsilateral retinofugal projections from the fight eye to the right hemisphere are ineffective for retrieval of information acquired via the left eye/right hemisphere system. Access from the left eye, both at training and at testing, would thus be confined to the right hemisphere. Therefore, it is conceivable that ablation of the right IMHV results in impaired memory when performing the task with the left eye (projecting to the right hemisphere), whereas removal of the left IMHV does not induce amnesia with either eye in use (note that whichever the eye used, the fight hyperstriatum can receive visual input via the thalamofugal ipsilateral or contralateral projections). It should be also noted that there are sex differences in the degree of the asymmetry in the thalamofugal projections, with males showing a greater degree of asymmetry than females. 36 Given the results of the present study, further research should be directed to analyse possible sex differences in the functional lateralization of the IMHV. One question to be addressed is the difference between other reports showing equivalent interocular transfer of the avoidance response between the two eye s y s t e m s 5'6'12 and our results showing a certain asymmetry of transfer, reported for a similar task. ~6 A possible explanation for such a discrepancy could be based on differences in the amount of time for which the chicks are monocularly deprived. In experimental conditions in which birds were binocularly pretrained and the eye-patches were allocated 10 min prior to training or testing there were no deleterious effects in direction or strength on interocular transfer. 5'6 The structural asymmetry in the thalamofugal pathway has been found to be dependent on lateralized light stimulation just prior to hatching, since the normal position of the embryo in the egg ensures that only the right eye is exposed to light. 39 The sensitive period for the asymmetrical effects of light on the development of these projections extends into the early posthatching period. 37 In our study, the longer period for which the chicks were monocularly deprived--including the period of pretraining trials---could have asymmetrically influenced functional development of the fibres from the undeprived eye to the ipsilateral hyperstriatum via the supraoptic decussation. As in the young chick these are almost exclusively from the fight eye, 39 it seems reasonable to suppose that only these fibres would be able to become functionally effective. Indeed, monocular deprivation during development has been shown to induce asymmetrical neurocbemical and structural changes in the visual Wulst (with an increase of activity on the side ipsilateral to the undeprived eye) in pigeons, 4"H which seems to be related to changes at the level of the supraoptic decussation. 9 However, these studies involved a long period (several weeks) of monocular deprivation. Our results suggest that a plastic change in the commis-
399
sural input could take place in 1 h of monocular occlusion. Despite the fact that our experimental conditions of monocular deprivation can modify the physiological processes occurring during learning in the intact animal, this manipulation has proved appropriate for addressing our main objective. Thus, the set of experiments performed combining unilateral IMHV lesions and monocular training/testing have indicated that the right IMHV is capable of acquiring the passive avoidance learning response (see also Ref. 3). Why, then, do lesions to the left part of the IMHV: (i) induce amnesia in binocularly trained animals if the right part can also participate in learning, and (ii) not impair learning in monocularly trained animals? An attempt to answer these questions requires the interrelation of two supportive bodies of data. First, it is now well established that the left and right eye systems of the young chick analyse and record different features of the environmentfl That is, the left eye system is mainly responsible for spatial orientation, whereas the right eye system is concerned with the selection of cues in order to assign stimuli into categories, including the identification of food. 2'33 It could be inferred from this behavioural specialization that, in the passive avoidance task, the right eye system should be the one concerned with the identification of the visual characteristics of the aversive bead. In fact, fight eye system advantage in visual discrimination tasks involving food as a goal has been reported both in chicks 26and in pigeons. 2° Second, in the intact day-old chick, the supraoptic decussation is still developing and is not yet believed to be functional? s If this commissure could become functional following monocular deprivation, the right eye would feed the hyperstriata in both hemispheres. Therefore, when the chick is binocularly trained in the passive avoidance task, the critical information about the bead's characteristics would be processed by the fight eye/left hemisphere. Even though the left eye/right hemisphere could be processing other kinds of information (for instance, spatial relations of the environment), such a model implies that only lesions to the left IMHV would disrupt object recognition, consequently rendering the chicks amnesic for the task. Thus, this explanation does not exclude the possibility that the fight IMHV is capable of sustaining learning processes. In fact, it is highly probable that this structure, although not indispensable for memory of the task, is processing certain information about the experience. 2 CONCLUSIONS
In summary, these findings suggest a relationship between lateralization of IMHV function and the visual asymmetries which occur at the behavioural (visually guided) and structural levels. Therefore, they open the question of whether the fight eye/left IMHV specialization for acquisition of the type of
400
C. SANDI et al.
learning involved in the passive avoidance response, is a response to an ontogenetic organisation of lateralized hemispheric functions, or whether it is exclusively a developmentally acquired contingency consequent on asymmetrical light stimulation in the egg. In fact, asymmetrical light input in the embryo not only determines asymmetry in the visual projections. 39 but also lateralization for control of certain behaviours, such as attack and copulation 34 and visual discrimination learningfl Interestingly, cerebral lateralization in humans has been suggested to be influenced by an asymmetrical prenatal development at the sensory level (ear and labyrinth) as a consequence of head position of the fetus, a2 In chicks, it has been suggested that light received by the right eye prior to hatching
could stimulate developmental processes in the left hemisphere in advance of the right. 35 A recent report has shown evidence of a lateralized location in the left hemisphere of visual memory function in pigeons. 46 Further work, involving the manipulation of light-exposure during the sensitive period (just before and after hatching), 37 could help understanding the mechanisms leading to lateralization of I M H V function for the acquisition of passive avoidance learning. Acknowledgements---C.S. was supported by a Postdoctoral
Fellowship from the CSIC (Spain). We thank Steve Waiters and Dawn Sadler for their consistent and careful maintenance of the chicks, and David Booth, Department of Psychology, University of Birmingham, for loan of the radiofrequency lesioning equipment.
REFERENCES
1. Ali S., Bullock S. and Rose S. P. R. (1988) Phosphorylation of synaptic proteins in chick forebrain: changes with development and passive avoidance training. J. Neurochem. 50, 1579-1587. 2. Andrew R. J. (1991) The nature of behavioural lateralization in the chick. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 536-554. Oxford University Press, New York. 3. Anokhin K. V., Mileusnic R., Shamakina I. Y. and Rose S. P. R. (1991) Effects of early experience on c-fos gene expression in the chick forebrain. Brain Res. 544, 101-107. 4. Bagnoli P., Casini G. and Alesci R. (1985) Changes in the avian visual Wulst following early monocular deprivation. In Brain Plasticity: Learning and Memory (eds Will B., Schmidtt P. and Dalrmyple-Alford J. C.), pp. 71 78. Plenum Press, New York. 5. Bell G. A. and Gibbs M. E. (1977) Unilateral storage of monocular engram in day-old chick. Brain Res. 124, 263-270. 6. Bell G. A. and Gibbs M. E. (1979) Interhemispheric engram transfer in chick. Neurosci. Left. 13, 163-168. 7. Boxer M. I. and Stanford D. (1985) Projections to the posterior visual hyperstriatal region of the chick: an HRP study. Expl Brain Res. 57, 494 498. 8. Burchuladze R., Potter J. and Rose S. P. R. (1990) Memory formation in the chick depends on membrane-bound protein kinase C. Brain Res. 535, 131-138. 9. Burkhalter A., Streit P., Bagnoli P., Vischer A., Henke H. and Cuenod M. (1982) Deprivation-induced functional modifications in the pigeon visual system. In Neuronal Plasticity and Memory Formation (eds Marsan C. A. and Matthies H.), pp. 477-485. Raven Press, New York. 10. Collins R. L. (1982) Toward an admissible genetic model for the inheritance of the degree and direction of asymmetry. In Lateralization in the Nervous System (eds Harnad S., Doty R. W., Goldstein L., Jaynes J. and Krauthamer G.), pp. 137-150. Academic Press, New York. 11. Cuenod M., Bagnoli P., Beaudet A., Burkhalter A., Henke H., Kniisel B. and Vischer A. (1981) Behavioral and biochemical changes induced by monocular deprivation in the pigeon. Doc. ophthal. Proc. Ser. 30, 230-236. 12. Cherkin A. (1970) Eye to eye transfer of an early response modification in chicks. Nature, Lond. 227, 1153. 13. Cowan W. M., Adamson L. and Powell T. P. S. (1961) An experimental study of the avian visual system. J. Anat., Lond. 95, 546-563. 14. Davies D. C. (1991) Lesion studies and the role of the IMHV in early learning. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 329-343. Oxford University Press, New York. 15. Davies D. C., Taylor D. A. and Johnson M. H. (1988) The effects of hyperstriatal lesions on one-trial passive avoidance learning in the chick. J. Neurosci. g, 4662-4666. 16. Gaston K. E. (1984) Interocular transfer of pattern discrimination learning in chicks. Brain Res. 310, 213-221. 17. Gilbert D. B., Patterson T. A. and Rose S. P. R. (1991) Dissociation of brain sites necessary for registration and storage of memory for a one-trial passive avoidance task in the chick. Behav. Neurosci. 105, 553-561. 18. Glick S. D. (1985) Cerebral Lateralization in Nonhuman Species. Academic Press, New York. 19. Glick S. D. and Shapiro R. M. (1984) Functional and neurochemical asymmetries. In Cerebral Dominance (eds Geschwind N. and Galaburda A. M.), pp. 147 166. Harvard University Press, Cambridge, MA. 20. Gfint/irk/in O. (1985) Lateralization of visually controlled behavior in pigeons. Physiol. Behav. 34, 575 577. 21. Giintfirkfin O. (1991) The functional organization of the avian visual system. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 92-105. Oxford University Press, New York. 22. Hellige J. B. (1990) Hemispheric asymmetry. A. Rev. Psychol. 41, 55-80. 23. Horn G. (1991) Imprinting and recognition memory: a review of neural mechanisms. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 219-261. Oxford University Press, New York. 24. Mason R. J. and Rose S. P. R. (1987) Lasting changes in spontaneous multi-unit activity in the chick brain following passive avoidance training. Neuroscience 21, 932-941. 25. McCabe B. J. (1991) Hemispheric asymmetry of learning-induced changes. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 262-276. Oxford University Press, New York. 26. Mench J. A. and Andrew R. K. (1986) Lateralization of a food search task in the domestic chick. Behav. neural Biol. 46, 107-114.
IMHV lesions and functional lateralization
401
27. Nottebohm F. (1979) Origins and mechanisms in the establishment of cerebral dominance. In Handbook o f Behavioral Neurobiology, Vol. 2 (ed. Gazzaniga M.), pp. 295-348. Plenum Press, New York. 28. Patel S. N. and Stewart M. G. (1988) Changes in the number and structure of dendritic spines, 25 h after passive avoidance training in the domestic chick, Gallus domesticus. Brain Res. 449, 34-46. 29. Patterson T. A., Alvarado M. C., Warner I. T., Bennet E. L. and Rosenzweig M. R. (1986) Memory stages and brain asymmetry in chick learning. Behav. Neurosci. 100, 856-865. 30. Patterson T. A., Gilbert D. B. and Rose S. P. R. (1990) Pre- and post-training lesions of the intermediate medial hyperstriatum ventrale and passive avoidance learning in the chick. Expl Brain Res. 80, 189-195. 31. Patterson T. A. and Rose S. P. R. (1992) Memory in chicks; multiple cues; separate locations. Behav. Neurasci. 106, 465-470. I 32. Previc F. H. (1991) A general theory concerning the prenatal origins of cerebral laterali~ation in humans. Psychol. Rev. 98, 299-334. 33. Rashid N. and Andrew R. J. (1989) Right hemisphere advantage for topographical orientation in the domestic chick. Neuropsychologia 27, 937-948. 34. Rogers L. J. (1982) Light experience and asymmetry of brain function in chickens. Nature 297, 223-225. 35. Rogers L. J. (1986) Lateralization of learning in chicks. Adv. Study Behav. 16, 147-1~9. 36. Rogers L. J. ( 1991) Development o f lateralization. In Neural and Behavioural Plasticity. i The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 507-535. Oxford University Press, New York. 37. Rogers L. J. and Bolden S. W. (1991) Light-dependent development and asymmetry of visual projections. Neurosci. Lett. 121, 63-67. 38. Rogers L. J., Robinson T. and Ehrlich D. (1986) Role of the supraoptic decussation in the development of asymmetry of brain function in the chicken. Devl Brain Res. 28, 33-39. 39. Rogers L. J. and Sink H. S. (1988) Transient asymmetry in the projections of the rostral thalamus to the visual hyperstriatum of the chicken, and reversal of its direction by light exposure. Expl Brain Res. 70, 378-384. 40. Rose S. P. R. (1991) Biochemical mechanisms involved in memory formation in the chick. In Neural and Behavioural Plasticity. The Use of the Domestic Chick as a Model (ed. Andrew R. J.), pp. 277-304. Oxford University Press, New York. 41. Rose S. P. R. (1991) How chicks make memories: the cellular cascade from c-fos to dendritic remodelling. Trends Neurosci. 14, 390-397. 42. Rose S. P. R. and Csillag A. (1985) Passive avoidance training results in lasting changes in deoxyglucose metabolism in left hemisphere regions of chick brain. Behav. neural Biol. 44, 315-324. 43. Sperry R. W., Gazzaniga M. S. and Bogen J. E. (1969) Interhemispheric relationships: the neocortical commisures, syndromes of hemispheric disconnection. In Handbook of Clinical Neurology, Iiol. 4: Disorders o f Speech, Perception, and Symbolic Behavior (eds Winken P. J. and Bruyn G. W.), pp. 145-153. Elsevier/North Holland and Biomedical Press, Amsterdam. 44. Stewart M. G. (1991) Changes in dendritic and synaptic structure in chick forebrain consequent on passive avoidance learning. In Neural and Behavioural Plasticity. The Use of the Domestic Chick as a Model (ed. Andrew R. J.), pp. 305-328. Oxford University Press, New York. 45. Stewart M. G., Rose S. P. R., King T. S., Gabbot P. L. A. and Bourne R. (1984) Hemispheric asymmetry of synapses in chick medial hyperstriatum ventrale following passive avoidance training: a stereological investigation. Devl Brain Res. 12, 261-269. 46. von Fersen L. and Giintfirkiin O. (1990) Visual memory lateralization in pigeons. Neuropsychologia 28, 1 7. 47. Youngren O. M. and Phillips R. E. (1978) A stereotaxic atlas of the 3 day old chick. J. comp. Neurol. 181, 567-600. 48. Zappia J. V. and Rogers L. J. (1983) Light experience during development affects asymmetry of forebrain function in chickens. Devl Brain Res. 11, 93 106. i
(Accepted 27 July 1992)
0306-4522/93$6.00+ 0.00 PergamonPress Ltd © 1992IBRO
VISUAL INPUT A N D LATERALIZATION OF BRAIN FUNCTION IN LEARNING IN THE CHICK C. SANDI,*T. A. PATTERSONand S. P. R. ROSE Brain and Behaviour Research Group, Department of Biology, The Open University, Milton Keynes MK7 6AA, U.K. Abstract--Several lines of evidence (biochemical,neuroanatomical, electrophysiological,and behavioural) have indicated a critical role for the intermediate medial hyperstriatum ventrale of the chick forebrain in the acquisition of a passive avoidance response. Previous lesion studies indicated that bilateral or left, but not right, pretraining intermediate medial hyperstriatum ventrale lesions interfere with the acquisition of this task. We have further analysed this asymmetrical involvement of the intermediate medial hyperstriatum ventrale by use of a monocular learning protocol and intermediate medial hyperstriatum ventrale lesions (sham, bilateral, or unilateral). The results indicated that there is interocular transfer of information of passive avoidance learning between the two eye systems, with a tendency to be more successful from the fight eye system to the left than in the opposite direction. As in binocular conditions, bilateral pretraining intermediate medial hyperstriatum ventrale lesions impair learning in monocularly trained animals. Unilateral lesions to either left or right monocularly trained experimental animals resulted in amnesia when they were made to the right intermediate medial hyperstriatum ventrale and the chicks were trained/tested with the left eye open. These results indicate that, although right intermediate medial hyperstriatum ventrale lesions do not result in amnesia in binocular animals, this region is capable of participating in memory acquisition processes. They also suggest a connection between lateralization of intermediate medial hyperstriatum ventrale function in passive avoidance learning and the behavioural and structural visual asymmetries known to occur in chicks.
The demonstration of functional hemispheric asymmetries in humans43 has stimulated research dealing with cerebral lateralization both in human and nonhuman species. 22 Although the phenomenon was initially considered specific to humans, during the past two decades a number of behavioural and biological asymmetries have been reported in other species) 8 Some of the functional asymmetries described in laboratory mammals 1°,~9are characterized by a marked stability over time for a specific individual but a varied distribution in direction and/or magnitude across the population. By contrast, most of the functional asymmetries observed in avian species have been found to be directionally consistent in the p o p u l a t i o n y °,27,37 resembling in this respect the well-established asymmetries of handedness and language in humans. In particular, the avian visual system provides a suitable tool for studying information processing in the separate brain hemispheres since the optic nerves virtually show a complete decussation at the chiasmatic level) 3 Using the domestic chick to study the neural mechanisms involved in early types of learning (e.g. imprinting, one-trial passive avoidance learning), a variety of asymmetric biochemical, morphological, *To whom correspondence should be addressed. Abbreviations: IMHV, intermediate medial hyperstriatum
ventrale; MeA, methylanthranilate.
and electrophysiological changes23,25,4°,~ have been found in different brain areas, and especially the intermediate medial hyperstriatum ventrale (IMHV), both in untrained and trained birds. In particular, following one trial passive avoidance training (in which chicks, which peck spontaneously at small beads, learn after a single trial to avoid a bead coated with a bitter substance), enhanced metabolic activity4~ and a consequent biochemical cascade have been shown to occur asymmetrically in the left IMHV. Ls,4~ Several measures of dendritic and synaptic morphology and number are specifically enhanced in the same region. 2s'44'45The strongest evidence for a lateralized involvement of the IMHV in learning stems from psychopharmacological and lesion studies. Thus, injection of amnestic agents, such as inhibitors of protein synthesis29 or protein kinase C 8 into the left, but not the right, IMHV results in amnesia for the passive avoidance. Left, but not right, pretraining IMHV lesions result in amnesia in chicks tested 3 h after training.3° However, post-training bilateral IMHV lesions do not interfere with retention of the task (though they do prevent discrimination of the bead based on colour), 31 suggesting that a memory trace once formed in the left IMHV is subsequently dispersed to other brain regions. However, not all changes are confined to the left IMHV; in the minutes to hours after training a variety of changes also occur bilaterally or in the right IMHV. 3,24,42
393
C. SANDI et al.
394
The aim of the present study was to use a m o n ocular learning protocol to analyse this asymmetrical involvement of the I M H V in the acquisition o f passive avoidance learning. Chicks have two m a j o r visual pathways, one projecting from the optic tectum to the ectostriatum a n d a n o t h e r projecting in a n asymmetric m a n n e r from the t h a l a m u s to the visual Wulst or hyperstriatum. 7,2~'39Visual information a b o u t the training experience is one of the m a j o r c o m p o n e n t s of passive avoidance learning, since trained chicks are able to discriminate between similar beads differing in some of their visual characteristics (for example, colour) 3t a n d the I M H V is considered to be a n area critically involved in certain kinds of visual recognition. 14 T h e present experiments were designed to study: (i) the role of the I M H V in interocular transfer of i n f o r m a t i o n of passive avoidance learning between the two eye/brain-related systems; a n d (ii) the possibility t h a t the right I M H V could participate in learning if visual i n f o r m a t i o n is forced to arrive at the right hemisphere by occluding the right eye.
EXPERIMENTAL PROCEDURES
were placed in pairs into pens (20 x 25 cm) with scattered chick crumbs. The pens were maintained at 28-30°C and were illuminated by an overhead red light (25 W). After the period of equilibration, the chicks were pretrained by three 10-s presentations of a small (2.5 mm) white bead, and then trained by a single presentation of a 4-mm bright chrome bead coated with methylanthranilate (MeA). Only chicks that pecked in at least two of the pretraining trials and, during training, evinced a clear disgust response (head shaking, bill wiping, etc.) were considered as experimental subjects (about 80% of chicks). When the experiments required changing the eye-patch from one eye to the other (Experiments I, 4), this was done 1 h after training. In all experiments, 3 h after training chicks were tested by being offered a dry chrome bead for 30 s, and their response (peck or avoid) was noted. To accumulate enough data to evaluate the effects of the treatments, several replications of each experiment (five to nine replications per experiment carried out throughout a nine-month period), including a balanced number of animals in each group, were made. The behavioural results were compared using the chi-squared test of independence. After training and testing in each experiment involving lesions, the chicks were killed and their brains were examined for sites of lesions using normal histological techniques. Both testing and histological examination were conducted by an experimenter blind to the past history of the chicks. Results from chicks with incorrectly placed lesions were excluded.
Animals Ross Chunky chicks (Gallus domesticus) of both sexes were hatched from commercially obtained eggs in communal brooders and maintained on a 12:12 h light
Surgical procedures Anaesthesia was induced by intraperitoneal injection of Equithesin (0.28 ml/100 g body weight). When narcosis became apparent, the chick was transferred to a stereotaxic instrument where its temperature was maintained by an electric underblanket. When the bird was fully anaesthetized, its cranium was secured in a modified small animal headholder, with the beak bar locked at 5 mm below and 11.5 mm anterior to the central axis of the ear bar. This orientation is similar to that used in Youngren and Phillip's47 atlas of the three-day-old chick brain. The scalp and underlying muscle insertions were deflected and small craniotomies were made in the skull directly over the electrode placement sites. After incision of the dura, the temperature-sensitive electrode was lowered under stereotaxic control. The coordinates used for the IMHV were 0.4 mm anterior to the ear bars, 0.9 mm lateral to the midline, and 1.7 mm ventral to the surface of the brain. Lesions were induced by radiofrequency using a Radionics radio frequency lesion generator (Model RFG4; Radionics, Inc., Burlington, MA). The temperature at the electrode tip was maintained at 60°C for 90 s. The electrode was removed when the temperature of the electrode probe returned to body temperature. The skull flap was closed and secured with warm bone wax. The scalp was sutured, and the chick was left to recover in a warm box for several hours. The chicks were then given water and allowed to recover in these boxes for several hours before training or testing. Sham control chicks underwent identical procedures to those of the lesioned animals, however, no current was passed through the electrode tip.
Training procedures The chicks were 24 h old at training. One hour before training, either the left (L) or the right (R) eye (cf. specific protocols) was occluded by means of a black paper patch, lightly stuck on to the down around the eye. Then, the chicks
RESULTS
Histology A histological reconstruction of a typical bilateral I M H V lesion is s h o w n in Fig. 1. Unilateral lesions were of the same size and at the same position o n the two sides of the brain as the bilateral lesions. The location and extent o f the lesions were similar a m o n g the different experimental groups as well as equivalent to those o b t a i n e d in previous studies involving lesions to this a r e a ) 7'3°
Experiment 1 Monocular learning, and testing the experienced or naive eye. A first experiment was carried o u t in order to establish the level of interocular transfer of the passive avoidance task u n d e r the m o n o c u l a r conditions chosen for these experiments. Twentyf o u r - h o u r old chicks were removed from the b r o o d e r s a n d either the left or right eye was covered with a n eye-patch. One h o u r after training, the eye-patch was carefully removed a n d a new one placed covering either the same eye again (in h a l f of the chicks in each m o n o c u l a r condition) or the opposite eye (in the other half o f the chicks in each m o n o c u l a r condition) (see Fig. 2). Since different sources of stress a r o u n d the training trial are k n o w n to interfere with the acquisition o f the task (Patterson, u n p u b l i s h e d observations), the schedule for allocating a n d c h a n g i n g the eye-patches (1 h prior a n d 1 h after the training trial, respectively) was selected to allow the chicks to h a b i t u a t e to the eye-patches a n d to reduce possible stress a r o u n d the training session. Pilot studies indicated t h a t eye-patches did not influence pecking in animals trained o n a water-coated bead.
IMHV lesions and functional lateralization HA
395 l-IV
A 3.5 HA
HV I-IV
A 6.0
@
HA
HV
A 3.0 HV
2.5 A 4.5 .V
A 4.0 Fig. I. Typical extent of bilateral IMHV lesions included in this study. Solid areas indicate total lysis of tissue, shaded areas indicate scar tissue degenerating neuropil. Unilateral lesions were of similar size and at the same position on the two sides of the brain as the bilateral lesions. A, archistriatum; E, ectostriatum; HA, hyperstriatum accesorium; HV, hyperstriatum ventrale; LPO, lobus parolfactorious; N, neostriatum; V, ventricle.
There was no difference between the two groups trained and tested with the same eye (~2 = 0.76, n.s.); i.e. both groups showed good levels of retention (left eye-patch t r a i n i n g + left eye-patch testing, LL = 82.1% of avoidance responses, n = 28; R R = 69.0%, n = 29). However, in the conditions involving testing with the untrained eye, the group trained with the left eye and tested with the right eye. (RL = 56.4%, n = 39) showed reduced avoidance as compared with the group trained with the right eye and tested with the left eye (LR = 77.5%, n = 40) (X2= 3.98, P = 0.05). Thus, transfer of information from the left
eye system to the right is slightly poorer than in the opposite direction (see Fig. 2).
Experiment 2 Effects of bilateral intermediate medial hyperstriatum ventrale lesions on monocular learning. Early monocular deprivation has been reported to induce different types of neural plasticity in the avian fore° brain. 4,n Therefore, before studying the effects of unilateral I M H V lesions (Experiments 3 and 4), we determined if the I M H V is necessary for the monocular acquisition of passive avoidance learning, as has
C. SANDl et al.
396 TRAIN
(~[~) ~
TEST
PREDICTIONS
RESULTS
n
% Avoidance
14 13 11 11
78.6 76.9 36.4* 45.4*
~D
, Recall
Recall
Sham lesion + left eye-patch Sham lesion + right eye-patch Bilat IMHV lesion + left eye-patch Bilat IMHV lesion + fight eye-patch
(~ [~o
,Recall
Recall
*P < 0.05 vs corresponding sham-lesioned group in the same eye-patch condition
I~[~
JRecall
Reduced Recell
Fig. 2. Protocol, predictions and results of Experiment 1 designed to establish the level of interocular transfer of the passive avoidance response when chicks are trained with one eye viewing, either left or right, and tested with the opposite eye. Open ellipse, eye open; filled ellipse, eye occluded; open circle, sham IMHV lesions; filled circle, genuine IMHV lesion. been shown to be the case in binocular experimental animals.~5,3° This was done by testing whether bilateral ablation of the I M H V prior to training interfered with the acquisition of passive avoidance learning in monocularly trained and tested chicks. At 18 h old, the chicks were given sham or bilateral I M H V lesions. One hour before training, chicks in each condition were divided into two experimental groups with either left or right eye covered, trained and tested 3 h later. The eye-patch was maintained in the same place until the end of the experiment (see Fig. 3). The results of this experiment are shown in Table 1. Just as the unoperated birds in Experiment 1, sham-operated animals displayed similar levels of avoidance with either eye occluded (n = 18 chicks per group, 3(2= 0.0, n.s.). However, pretraining bilateral I M H V lesions impaired the acquisition of passive avoidance learning in monocularly, left (n = 15, Z2=4.90, P <0.005) or right (n = 2 0 , Z2=5.39, TRAIN
Table 1. Effects of sham or bilateral intermediate medial hyperstriatum ventrale lesions on passive avoidance learning in monocular, left or right trained and tested chicks
TEST
PREDICTIONS
P < 0.025), trained and tested animals. These results indicate that, as is the case in the binocular conditions, ~5'3° an intact I M H V is necessary for the successful acquisition of passive avoidance learning using only one eye.
Experiment 3 Effects o f unilateral intermediate medial hyperstriatum ventrale lesions on monocular learning using the same eye at training and at testing. Having established that bilateral lesions to the I M H V result in amnesia for monocular learning, the next question was whether in a monocular learning situation there would be a lateralized involvement of the I M H V depending on the eye used to perform the task. Chicks were given unilateral, left or fight, lesions and trained either with the left or with the right eye open. Because left but not right I M H V lesions result in disruption of learning in birds viewing the bead binocularly, 3° we made the following predictions (see Fig. 4): (i) left I M H V lesions given to chicks trained with the fight eye open (left eye-patch) should render the chicks amnesic for the task; (ii) lesions performed in the I M H V ipsilateral to the open eye, either left or right (and so contralateral to the eye-patch), would not have any effect on learning, unless the left I M H V is critical for the associative processing related to the different components (i.e. visual, gustatory, emotional) of the task, in which case left I M H V lesions could
RESULTS TRAIN
TEST
PREDICTIONS
RESULTS
(Amnesia?)
(~)
, Q~[~)
, Amnesia
Amnesic
Fig. 3. Protocol, predictions and results of Experiment 2 designed to determine whether the IMHV is necessary for the monocular acquisition of passive avoidance learning. Lesions were made pretraining on the day the chicks hatched, the chicks were trained the following day and tested 3 h subsequently. Open ellipse, eye open; filled ellipse, eye occluded; open circle, sham IMHV lesions; filled circle, genuine IMHV lesion.
Fig. 4. Protocol, predictions and results of Experiment 3 designed to determine whether there is a lateralized involvement of the IMHV on passive avoidance learning depending on the eye used to perform the task. Open ellipse, eye open; filled ellipse, eye occluded; open circle, sham IMHV lesions; filled circle, genuine IMHV lesion.
IMHV lesions and functional lateralization Table 2. Effects of unilateral left or right intermediate medial hyperstriatum ventrale lesions in chicks trained/ tested with one eye, left or right, occluded Left IMHV lesion + left eye-patch Left IMHV lesion + right eye-patch Right IMHV lesion + left eye-patch Right IMHV lesion + right eye-patch
n
% Avoidance
23 19 19 19
65.2 78.9 78.9 42.1"
*P < 0.05 vs left IMHV-lesioned + right eye-patch and vs right IMHV lesioned + left eye-patch groups. also impair learning in left eye-trained/tested chicks; and (iii) if the right IMHV is able to take over the role of the left IMHV in learning acquisition processes, when the visual information is made to arrive mainly to the right hemisphere by occluding the right eye, right IMHV lesions would result in amnesia in chicks trained/tested with the left eye (right eye-patch); but if this take-over is not possible, the chicks would perform successfully in the task. At 18 h old, the chicks were given one of two types of lesion: either (i) a lesion in the left IMHV and a sham lesion in the right IMHV (left IMHV lesion group); or (ii) a lesion in the right IMHV and a sham lesion in the left IMHV (right IMHV lesion group). All other procedures were similar to those of Experiment 2. The results of this Experiment are shown in Table 2. When chicks were given left IMHV lesions, there were no significant differences between the right eye- (n = 23) and left eye- (n = 19) trained groups (Z2= 0.41, n.s.); i.e. both groups showed high levels of avoidance at retention. However, when the lesions were made to the right IMHV, chicks trained with the left eye (n = 19) displayed significantly lower levels of avoidance than chicks trained with the right eye (n = 19) which displayed a good retention level (~2 = 3.96, P < 0.05). In addition, the right IMHVlesioned group in the left-eye open condition also differed significantly from the left IMHV-lesioned group in the same eye viewing condition (X2 = 3.96, P < 0.05), but it was no different from the left IMHV lesion plus right-eye open group (X2= 2.24). There were no differences between the performance of the right IMHV lesion and left eye-patch group and left IMHV-lesioned animals in either monocular training/ testing condition, right-eye open (Z2= 0.41, n.s.) or left-eye open (~(2= 0.00, n.s.) (see Fig. 4). Therefore, our first prediction about an amnesic effect for left IMHV lesions given to chicks performing with the right eye seeing (left eye-patch) was not supported by the present findings. There was, however, supporting evidence for the second prediction in that lesions in the IMHV ipsilateral to the eye open, regardless of which eye it is, would not have any effect on learning. The fact that performance of chicks with left IMHV lesions and performing with the left eye open (right eye-patch) (main visual information arriving to the right hemisphere) was not impaired seems to indicate that the right IMHV could be critically involved in the acquisition processes of the
397
task when the information is forced to be, at least primarily, processed in the right hemisphere. The strongest evidence for such a role comes from the findings related to the third prediction, showing that when information is forced to arrive at the fight hemisphere (right eye-patch group), right IMHV lesions result in amnesia. In summary, these striking results indicated that in our monocular learning protocol, left IMHV lesions did not interfere with learning regardless of the eye used, but fight IMHV lesions impaired performance in the left eye open group. Thus, what it is revealed is a lateralized effect of IMHV lesions opposite to that found in binocular learning) ° Could this result be a consequence of the known asymmetry in the visual pathways (see Discussion)? To test this hypothesis, we reasoned that if unilateral, left or right, IMHVlesioned chicks were monocularly trained under the same conditions as in Experiment 3 but were tested for retention using the untrained eye, the results should reflect this asymmetry in the visual projections (Experiment 4).
Experiment 4 Effects of unilateral intermediate medial hyperstriatum ventrale lesions on monocular learning using a different eye at training and at testing. Chicks given unilateral IMHV lesions were monocularly trained and then tested using the opposite eye. Based on the results obtained in the previous experiments the predictions were (see Fig. 5): (i) left IMHV lesions would not interfere with performance of chicks trained with the right eye and tested with the left eye (left eyepatch training + right eye-patch testing); (ii) chicks given left IMHV lesions and performing at training with the left eye and at testing with the right eye (right eye-patch training + left eye-patch testing), may show impairment in retention since interocular transfer from the left to the right eye was already shown to be slightly poorer in non-lesioned chicks in Experiment 1; (iii) if the lesions are given to the right IMHV and the chicks are trained with the right eye, in accordance with Experiment 3, the chicks should successfully acquire the task; however, if they are tested for retention with the left eye (left eye-patch training + right eye-patch testing) they should be now amnesic since the left eye is not well connected with the left hemisphere (see Discussion; also see Refs 7, 39); and (iv) since chicks given right IMHV lesions and trained with the left eye (which feeds to the right hemisphere) were shown to be amnesic (Experiment 3), if they are now tested with the right eye (right eye-patch training + left eye-patch testing), they should also be amnesic since they would have not learnt the task. Chicks were given unilateral IMHV lesions as in Experiment 3. As before, birds in each lesion condition were divided into experimental groups with either left or right eye occluded 1 h before training. There were therefore four groups including all the combinations between lesion x eye-patch conditions. By contrast to
C. SANDI et al.
398 TRAIN
TEST
,
PREDICTIONS
o
Recall
RESULTS
Recall
IMHV lesions, in chicks performing monocularly in a passive avoidance task, are asymmetric (see Fig. 5). They also confirm the result of Experiment 1 and Gaston ~6 about a mild detrimental effect for interocular transfer when the information learned with the left eye has to be retrieved with the fight eye. DISCUSSION
The left IMHV has been regarded as a structure necessary for successful acquisition of passive avoidance learning. On the basis of a wide body of data showing consistent asymmetrical learning-induced Fig. 5. Protocol, predictions and results of Experiment 4 biochemical and cellular changes in the IMHV in designed to determine whether there is an asymmetrical effect of the involvement of the IMHV on retention for chicks exposed to a passive avoidance task (see passive avoidance learning depending on the eye used at Introduction), and our previous finding that left, but training and at testing. Open ellipse, eye open; filled ellipse, not fight, IMHV lesions interfere with acquisition of eye occluded; open circle, sham IMHV lesions; filled circle, the avoidance response) ° Although a set of subsegenuine IMHV lesion. quent lesion experiments suggested a possible role of the right IMHV in distributing the information Experiment 3, 1 h after training the eye-patches were acquired from the left IMHV to other brain areas, ~7 transferred to the opposite eye. The results of this the physiological basis for such a lateralization of experiment are shown in Table 3. As can be seen, IMHV function in learning remained obscure. It chicks receiving left IMHV lesions and performing in could be that the left IMHV itself is uniquely capable the right-eye open at training + left-eye open at test- of learning-induced plasticity, or, alternatively, that ing conditions (n = 19) showed good levels of avoid- it assumed this role as a consequence of prior aspects ance. Left IMHV-lesioned chicks in the left-eye open of hemispheric asymmetry such as, for example, at training + right-eye open at testing group (n = 13) visual processing. Under our experimental monocular conditions, showed a slight but non-significant trend to reduced avoidance scores by comparison with their IMHV- there was interocular transfer of information of passive lesioned counterparts in the opposite eye viewing avoidance learning, although this transfer tended to condition (Z2= 1.46, P < 0.25). When the lesions be less successful from the left eye to the right eye were given to the right IMHV, both groups showed than in the opposite direction (Experiments 1 and 4). low levels of avoidance as compared with the "left Given that bilateral IMHV lesions were shown to interfere with learning in monocularly trained chicks IMHV lesion and right-eye open at training + left-eye open at testing" group showing normal retention (Experiment 2), the lack of effect of left IMHV lesions observed under monocular conditions (Experiment 3) levels (right IMHV lesion, left eye-patch training + fight eye-patch testing: n = 21, Z2 =4.99, P < 0.03; could be interpreted as the consequence of an involveright IMHV lesion, right eye-patch training + left ment of the fight IMHV in learning processes. The eye-patch testing: n = 19, X2= 3.84, P = 0.05), but fact that only fight IMHV lesions disrupted learning there was no difference between the two monocular when chicks performed with the left eye, but left IMHV lesions were ineffective even when chicks conditions analysed (X2= 0.05, n.s.). Therefore, these results confirm the predictions performed with the contralateral, right eye suggests made for each experimental condition on the basis of a possible relation between the lateralised effects the hypothesis that the laterality effects of unilateral of IMHV lesions and the asymmetry of visual projections. Stronger support for this interpretation was Table 3. Effects of unilateral left or right intermediate obtained when chicks trained with the fight eye, which medial hyperstriatum ventrale lesionson avoidance scores in are known to acquire the task successfully either with chicks trained with one eye and tested with the opposite one a left or a right IMHV lesion (Experiment 3), were n % Avoidance tested for retention with the left eye (Experiment 4). Under such circumstances, chicks with a lesion in the Left IMHV lesion + left eye-patch 19 73.7 training + right eye-patch testing fight, but not in the left, IMHV showed amnesia for the task. Left IMHV lesion + right eye-patch 13 46.2 It is a peculiar feature of the chick, during the first training + left eye-patch testing three weeks of life, that there is an asymmetry in the Right IMHV lesion + left eye-patch 21 33.3* visual thalamofugal projections; that is, the fight eye training + right eye-patch testing projects to both left and fight hemispheres, whereas Right IMHV lesion + right eye-patch 19 36.8* the left eye is connected almost only with the fight training + left eye-patch testing *P < 0.05 vs corresponding eye-patch group in the left hemisphere. 7,39 This asymmetry suggests that one explanation for our results is that when the right eye IMHV lesion condition.
IMHV lesions and functional lateralization is monocularly trained, information about the task will be accessible for processing in both hemispheres but that the ipsilateral retinofugal projections from the fight eye to the right hemisphere are ineffective for retrieval of information acquired via the left eye/right hemisphere system. Access from the left eye, both at training and at testing, would thus be confined to the right hemisphere. Therefore, it is conceivable that ablation of the right IMHV results in impaired memory when performing the task with the left eye (projecting to the right hemisphere), whereas removal of the left IMHV does not induce amnesia with either eye in use (note that whichever the eye used, the fight hyperstriatum can receive visual input via the thalamofugal ipsilateral or contralateral projections). It should be also noted that there are sex differences in the degree of the asymmetry in the thalamofugal projections, with males showing a greater degree of asymmetry than females. 36 Given the results of the present study, further research should be directed to analyse possible sex differences in the functional lateralization of the IMHV. One question to be addressed is the difference between other reports showing equivalent interocular transfer of the avoidance response between the two eye s y s t e m s 5'6'12 and our results showing a certain asymmetry of transfer, reported for a similar task. ~6 A possible explanation for such a discrepancy could be based on differences in the amount of time for which the chicks are monocularly deprived. In experimental conditions in which birds were binocularly pretrained and the eye-patches were allocated 10 min prior to training or testing there were no deleterious effects in direction or strength on interocular transfer. 5'6 The structural asymmetry in the thalamofugal pathway has been found to be dependent on lateralized light stimulation just prior to hatching, since the normal position of the embryo in the egg ensures that only the right eye is exposed to light. 39 The sensitive period for the asymmetrical effects of light on the development of these projections extends into the early posthatching period. 37 In our study, the longer period for which the chicks were monocularly deprived--including the period of pretraining trials---could have asymmetrically influenced functional development of the fibres from the undeprived eye to the ipsilateral hyperstriatum via the supraoptic decussation. As in the young chick these are almost exclusively from the fight eye, 39 it seems reasonable to suppose that only these fibres would be able to become functionally effective. Indeed, monocular deprivation during development has been shown to induce asymmetrical neurocbemical and structural changes in the visual Wulst (with an increase of activity on the side ipsilateral to the undeprived eye) in pigeons, 4"H which seems to be related to changes at the level of the supraoptic decussation. 9 However, these studies involved a long period (several weeks) of monocular deprivation. Our results suggest that a plastic change in the commis-
399
sural input could take place in 1 h of monocular occlusion. Despite the fact that our experimental conditions of monocular deprivation can modify the physiological processes occurring during learning in the intact animal, this manipulation has proved appropriate for addressing our main objective. Thus, the set of experiments performed combining unilateral IMHV lesions and monocular training/testing have indicated that the right IMHV is capable of acquiring the passive avoidance learning response (see also Ref. 3). Why, then, do lesions to the left part of the IMHV: (i) induce amnesia in binocularly trained animals if the right part can also participate in learning, and (ii) not impair learning in monocularly trained animals? An attempt to answer these questions requires the interrelation of two supportive bodies of data. First, it is now well established that the left and right eye systems of the young chick analyse and record different features of the environmentfl That is, the left eye system is mainly responsible for spatial orientation, whereas the right eye system is concerned with the selection of cues in order to assign stimuli into categories, including the identification of food. 2'33 It could be inferred from this behavioural specialization that, in the passive avoidance task, the right eye system should be the one concerned with the identification of the visual characteristics of the aversive bead. In fact, fight eye system advantage in visual discrimination tasks involving food as a goal has been reported both in chicks 26and in pigeons. 2° Second, in the intact day-old chick, the supraoptic decussation is still developing and is not yet believed to be functional? s If this commissure could become functional following monocular deprivation, the right eye would feed the hyperstriata in both hemispheres. Therefore, when the chick is binocularly trained in the passive avoidance task, the critical information about the bead's characteristics would be processed by the fight eye/left hemisphere. Even though the left eye/right hemisphere could be processing other kinds of information (for instance, spatial relations of the environment), such a model implies that only lesions to the left IMHV would disrupt object recognition, consequently rendering the chicks amnesic for the task. Thus, this explanation does not exclude the possibility that the fight IMHV is capable of sustaining learning processes. In fact, it is highly probable that this structure, although not indispensable for memory of the task, is processing certain information about the experience. 2 CONCLUSIONS
In summary, these findings suggest a relationship between lateralization of IMHV function and the visual asymmetries which occur at the behavioural (visually guided) and structural levels. Therefore, they open the question of whether the fight eye/left IMHV specialization for acquisition of the type of
400
C. SANDI et al.
learning involved in the passive avoidance response, is a response to an ontogenetic organisation of lateralized hemispheric functions, or whether it is exclusively a developmentally acquired contingency consequent on asymmetrical light stimulation in the egg. In fact, asymmetrical light input in the embryo not only determines asymmetry in the visual projections. 39 but also lateralization for control of certain behaviours, such as attack and copulation 34 and visual discrimination learningfl Interestingly, cerebral lateralization in humans has been suggested to be influenced by an asymmetrical prenatal development at the sensory level (ear and labyrinth) as a consequence of head position of the fetus, a2 In chicks, it has been suggested that light received by the right eye prior to hatching
could stimulate developmental processes in the left hemisphere in advance of the right. 35 A recent report has shown evidence of a lateralized location in the left hemisphere of visual memory function in pigeons. 46 Further work, involving the manipulation of light-exposure during the sensitive period (just before and after hatching), 37 could help understanding the mechanisms leading to lateralization of I M H V function for the acquisition of passive avoidance learning. Acknowledgements---C.S. was supported by a Postdoctoral
Fellowship from the CSIC (Spain). We thank Steve Waiters and Dawn Sadler for their consistent and careful maintenance of the chicks, and David Booth, Department of Psychology, University of Birmingham, for loan of the radiofrequency lesioning equipment.
REFERENCES
1. Ali S., Bullock S. and Rose S. P. R. (1988) Phosphorylation of synaptic proteins in chick forebrain: changes with development and passive avoidance training. J. Neurochem. 50, 1579-1587. 2. Andrew R. J. (1991) The nature of behavioural lateralization in the chick. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 536-554. Oxford University Press, New York. 3. Anokhin K. V., Mileusnic R., Shamakina I. Y. and Rose S. P. R. (1991) Effects of early experience on c-fos gene expression in the chick forebrain. Brain Res. 544, 101-107. 4. Bagnoli P., Casini G. and Alesci R. (1985) Changes in the avian visual Wulst following early monocular deprivation. In Brain Plasticity: Learning and Memory (eds Will B., Schmidtt P. and Dalrmyple-Alford J. C.), pp. 71 78. Plenum Press, New York. 5. Bell G. A. and Gibbs M. E. (1977) Unilateral storage of monocular engram in day-old chick. Brain Res. 124, 263-270. 6. Bell G. A. and Gibbs M. E. (1979) Interhemispheric engram transfer in chick. Neurosci. Left. 13, 163-168. 7. Boxer M. I. and Stanford D. (1985) Projections to the posterior visual hyperstriatal region of the chick: an HRP study. Expl Brain Res. 57, 494 498. 8. Burchuladze R., Potter J. and Rose S. P. R. (1990) Memory formation in the chick depends on membrane-bound protein kinase C. Brain Res. 535, 131-138. 9. Burkhalter A., Streit P., Bagnoli P., Vischer A., Henke H. and Cuenod M. (1982) Deprivation-induced functional modifications in the pigeon visual system. In Neuronal Plasticity and Memory Formation (eds Marsan C. A. and Matthies H.), pp. 477-485. Raven Press, New York. 10. Collins R. L. (1982) Toward an admissible genetic model for the inheritance of the degree and direction of asymmetry. In Lateralization in the Nervous System (eds Harnad S., Doty R. W., Goldstein L., Jaynes J. and Krauthamer G.), pp. 137-150. Academic Press, New York. 11. Cuenod M., Bagnoli P., Beaudet A., Burkhalter A., Henke H., Kniisel B. and Vischer A. (1981) Behavioral and biochemical changes induced by monocular deprivation in the pigeon. Doc. ophthal. Proc. Ser. 30, 230-236. 12. Cherkin A. (1970) Eye to eye transfer of an early response modification in chicks. Nature, Lond. 227, 1153. 13. Cowan W. M., Adamson L. and Powell T. P. S. (1961) An experimental study of the avian visual system. J. Anat., Lond. 95, 546-563. 14. Davies D. C. (1991) Lesion studies and the role of the IMHV in early learning. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 329-343. Oxford University Press, New York. 15. Davies D. C., Taylor D. A. and Johnson M. H. (1988) The effects of hyperstriatal lesions on one-trial passive avoidance learning in the chick. J. Neurosci. g, 4662-4666. 16. Gaston K. E. (1984) Interocular transfer of pattern discrimination learning in chicks. Brain Res. 310, 213-221. 17. Gilbert D. B., Patterson T. A. and Rose S. P. R. (1991) Dissociation of brain sites necessary for registration and storage of memory for a one-trial passive avoidance task in the chick. Behav. Neurosci. 105, 553-561. 18. Glick S. D. (1985) Cerebral Lateralization in Nonhuman Species. Academic Press, New York. 19. Glick S. D. and Shapiro R. M. (1984) Functional and neurochemical asymmetries. In Cerebral Dominance (eds Geschwind N. and Galaburda A. M.), pp. 147 166. Harvard University Press, Cambridge, MA. 20. Gfint/irk/in O. (1985) Lateralization of visually controlled behavior in pigeons. Physiol. Behav. 34, 575 577. 21. Giintfirkfin O. (1991) The functional organization of the avian visual system. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 92-105. Oxford University Press, New York. 22. Hellige J. B. (1990) Hemispheric asymmetry. A. Rev. Psychol. 41, 55-80. 23. Horn G. (1991) Imprinting and recognition memory: a review of neural mechanisms. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 219-261. Oxford University Press, New York. 24. Mason R. J. and Rose S. P. R. (1987) Lasting changes in spontaneous multi-unit activity in the chick brain following passive avoidance training. Neuroscience 21, 932-941. 25. McCabe B. J. (1991) Hemispheric asymmetry of learning-induced changes. In Neural and Behavioural Plasticity. The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 262-276. Oxford University Press, New York. 26. Mench J. A. and Andrew R. K. (1986) Lateralization of a food search task in the domestic chick. Behav. neural Biol. 46, 107-114.
IMHV lesions and functional lateralization
401
27. Nottebohm F. (1979) Origins and mechanisms in the establishment of cerebral dominance. In Handbook o f Behavioral Neurobiology, Vol. 2 (ed. Gazzaniga M.), pp. 295-348. Plenum Press, New York. 28. Patel S. N. and Stewart M. G. (1988) Changes in the number and structure of dendritic spines, 25 h after passive avoidance training in the domestic chick, Gallus domesticus. Brain Res. 449, 34-46. 29. Patterson T. A., Alvarado M. C., Warner I. T., Bennet E. L. and Rosenzweig M. R. (1986) Memory stages and brain asymmetry in chick learning. Behav. Neurosci. 100, 856-865. 30. Patterson T. A., Gilbert D. B. and Rose S. P. R. (1990) Pre- and post-training lesions of the intermediate medial hyperstriatum ventrale and passive avoidance learning in the chick. Expl Brain Res. 80, 189-195. 31. Patterson T. A. and Rose S. P. R. (1992) Memory in chicks; multiple cues; separate locations. Behav. Neurasci. 106, 465-470. I 32. Previc F. H. (1991) A general theory concerning the prenatal origins of cerebral laterali~ation in humans. Psychol. Rev. 98, 299-334. 33. Rashid N. and Andrew R. J. (1989) Right hemisphere advantage for topographical orientation in the domestic chick. Neuropsychologia 27, 937-948. 34. Rogers L. J. (1982) Light experience and asymmetry of brain function in chickens. Nature 297, 223-225. 35. Rogers L. J. (1986) Lateralization of learning in chicks. Adv. Study Behav. 16, 147-1~9. 36. Rogers L. J. ( 1991) Development o f lateralization. In Neural and Behavioural Plasticity. i The Use o f the Domestic Chick as a Model (ed. Andrew R. J.), pp. 507-535. Oxford University Press, New York. 37. Rogers L. J. and Bolden S. W. (1991) Light-dependent development and asymmetry of visual projections. Neurosci. Lett. 121, 63-67. 38. Rogers L. J., Robinson T. and Ehrlich D. (1986) Role of the supraoptic decussation in the development of asymmetry of brain function in the chicken. Devl Brain Res. 28, 33-39. 39. Rogers L. J. and Sink H. S. (1988) Transient asymmetry in the projections of the rostral thalamus to the visual hyperstriatum of the chicken, and reversal of its direction by light exposure. Expl Brain Res. 70, 378-384. 40. Rose S. P. R. (1991) Biochemical mechanisms involved in memory formation in the chick. In Neural and Behavioural Plasticity. The Use of the Domestic Chick as a Model (ed. Andrew R. J.), pp. 277-304. Oxford University Press, New York. 41. Rose S. P. R. (1991) How chicks make memories: the cellular cascade from c-fos to dendritic remodelling. Trends Neurosci. 14, 390-397. 42. Rose S. P. R. and Csillag A. (1985) Passive avoidance training results in lasting changes in deoxyglucose metabolism in left hemisphere regions of chick brain. Behav. neural Biol. 44, 315-324. 43. Sperry R. W., Gazzaniga M. S. and Bogen J. E. (1969) Interhemispheric relationships: the neocortical commisures, syndromes of hemispheric disconnection. In Handbook of Clinical Neurology, Iiol. 4: Disorders o f Speech, Perception, and Symbolic Behavior (eds Winken P. J. and Bruyn G. W.), pp. 145-153. Elsevier/North Holland and Biomedical Press, Amsterdam. 44. Stewart M. G. (1991) Changes in dendritic and synaptic structure in chick forebrain consequent on passive avoidance learning. In Neural and Behavioural Plasticity. The Use of the Domestic Chick as a Model (ed. Andrew R. J.), pp. 305-328. Oxford University Press, New York. 45. Stewart M. G., Rose S. P. R., King T. S., Gabbot P. L. A. and Bourne R. (1984) Hemispheric asymmetry of synapses in chick medial hyperstriatum ventrale following passive avoidance training: a stereological investigation. Devl Brain Res. 12, 261-269. 46. von Fersen L. and Giintfirkiin O. (1990) Visual memory lateralization in pigeons. Neuropsychologia 28, 1 7. 47. Youngren O. M. and Phillips R. E. (1978) A stereotaxic atlas of the 3 day old chick. J. comp. Neurol. 181, 567-600. 48. Zappia J. V. and Rogers L. J. (1983) Light experience during development affects asymmetry of forebrain function in chickens. Devl Brain Res. 11, 93 106. i
(Accepted 27 July 1992)