- Email: [email protected]
Effects of telencephalic ablation on shoaling behavior in goldfish
Physiology & Behavior 81 (2004) 141 – 148
Effects of telencephalic ablation on shoaling behavior in goldfish Kazutaka Shinozuka*, Shigeru Watanabe Department of Psychology, Keio University, Mita 2-15-45, Minato-ku, Tokyo, Japan Received 28 February 2003; received in revised form 10 October 2003; accepted 20 January 2004
Abstract The role of the telencephalon in the shoaling behavior of the goldfish, Carassius auratus, was investigated. Experiments were carried out in a tank divided into three compartments. Subjects were introduced into the center compartment of the tank. In the activity phase, subjects swam alone, and swimming distance was used as an index of activity. In the shoaling behavior phase, a stimulus fish was introduced into one of the side compartments, and the time spent near the side compartment by the subject was used as index of shoaling behavior. After these measurements were made, visual and motor abilities were examined using the optomotor response. Subjects then received surgery and the same procedure was repeated. In Experiment 1, the effects of total ablation of the telencephalon and a section of the olfactory tract (OlT) were examined. The ablation group exhibited reduced activity and shoaling behavior compared with the sham and OlT group. In Experiment 2, the role of the dorsal part of the telencephalon was examined after damaging the dorsomedial and dorsolateral telencephalon. Lesions in either portion had no effect and no simple visual or motor deficits were seen. These results suggest that the ventral part of the telencephalon mediates shoaling behavior. D 2004 Elsevier Inc. All rights reserved. Keywords: Goldfish; Shoaling behavior; Social behavior; Telencephalon
1. Introduction Numerous studies have shown that many factors affect shoaling behavior in fish, including number of members [12], body size [13], body coloration [18], hunger [29], and foraging behavior [26]. Shoaling has advantages (e.g., predator avoidance and food discovery) and disadvantages (e.g., competition for foraging). Fish tend to choose a shoal of fish with similar body size, because large fish in a shoal of small members may be striking and the risk of predation would be high. On the other hand, small fish in a shoal with large members may experience competition for food. Hungry fish tend to choose a small group rather than a large group because competition for food is reduced in the small group, although the risk of being preyed upon is increased [29]. Although many studies have investigated factors affecting shoaling behavior, the region of the brain involved in shoaling behavior is not known. The teleost telencephalon, which was once considered an olfactory system, is now known to be involved in more complex functions, such as * Corresponding author. Tel.: +81-354-433896; fax: +81-354-433897. E-mail address: [email protected] (K. Shinozuka). 0031-9384/$ – see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2004.01.005
reproductive behavior and learning. For example, telencephalon-ablated fish show reduced spawning behavior [6,10, 11]. Telencephalic ablation also impairs some types of learning, including habituation [16], complex operant tasks [8], avoidance tasks [31], and spatial learning [30]. Anatomically, the teleost telencephalon can be divided into two parts—the area dorsalis telencephali (D) and the area ventralis telencephali (V). Although differences exist between species [23], in general, the dorsal part consists of five substructures: the pars medialis (Dm), the pars dorsalis (Dd), the pars lateralis (Dl), the pars centralis (Dc), and the pars posterior (Dp). The ventral part consists of seven substructures: the pars ventralis (Vv), the pars dorsalis (Vd), the pars lateralis (Vl), the pars supracommissuralis (Vs), the pars postocommissuralis (Vp), the pars intermedia (Vi), and the nucleus entopeduncularis (E) [35]. Homology of these substructures to the brains of land vertebrates cannot be easily determined because the teleost brain undergoes a different process of development, called eversion. As a result of eversion, the pallium of the teleost telencephalon reverses mediolaterally compared with land vertebrates [35]. On the basis of neuronal connections, morphology, and histochemistry, the substructures Vs and area ventralis telencephali pars centralis (Vc) are considered homologous to
142
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
the basal amygdala of land vertebrates, and Vp, Vi, and nucleus taenia (NT) are considered homologous to the pallial amygdala [22]. More recently, however, it was proposed that Dm is homologous to the pallial amygdala [3]. Thus, it is difficult to draw homologies between the parts of the teleost telencephalon and their counterparts in other vertebrates. Using behavioral studies, some researchers claim there is homology between the teleost telencephalon and the mammalian limbic system [8,15,28]. Although most of these studies examined the effects of total ablation of the telencephalon, a few investigated the effects of partial lesions. For example, Dl lesions impaired performance in spatial learning tasks, which suggests that Dl is homologous to the hippocampus of land vertebrates. Dm lesions impaired performance on the shuttle avoidance task, indicating that the effect was caused by disruption of emotional memory, which suggests that Dm could be homologous to the pallial amygdala of land vertebrates [28]. Lesions in the Vs and posterior part of Vv (pVv) regions blocked the initiation of spawning behavior, suggesting homology between these regions and the medial amygdala of land vertebrates [14]. The purpose of the present study was to investigate whether the telencephalon mediates shoaling behavior in goldfish. Goldfish have been extensively used for behavioral studies, and they do shoal [17,25,27,33]. The effects of ablation of the total telencephalon were examined in Experiment 1, and the effects of partial lesions in the dorsal part of the telencephalon were examined in Experiment 2.
2. Experiment 1 2.1. Methods 2.1.1. Animals Twenty-one goldfish (Carassius auratus) were obtained from a local dealer. Their mean body length was 8.0 cm (range, 7.4– 9.0 cm). One randomly selected fish was used as a stimulus fish and the rest were used as subjects. The fish were housed in a Plexiglas tank of filtered water. Experiments were not begun until at least 7 days after the fish arrived. The fish were fed daily on commercial goldfish food. 2.1.2. Apparatus The tank in which experiments were carried out was 60 cm wide, 30 cm long and 36 cm deep, and was divided into three areas by transparent glass panels. The center area was 30 cm wide and 30 cm long. The side areas were 15 cm wide and 30 cm long. The water depth was 5 cm. A CCD camera (XC-711, SONY) set above the tank sent its signal to the Chromascan (OKK) video tracker (G220, OKK) that measured the position of the goldfish by color detection. An IBM PC/AT compatible computer recorded the data.
To examine the optomotor response, a circular tank (32 cm diameter and 20 cm deep) was used. A glass cylinder (5.5 cm diameter) was set in the center of the tank. The depth of the water in the tank was 5 cm. A drum (34 cm diameter and 21 cm high) painted with vertical black and white lines (3.2 cm wide) was placed around the tank and rotated clockwise or counterclockwise 15 times/min. During this task, fish tend to chase the rotation of the drum, allowing visual and motor ability to be measured by their optomotor response. 2.1.3. Procedure The experiment consisted of seven phases: (1) presurgery baseline activity, (2) presurgery shoaling behavior, (3) presurgery optomotor response, (4) surgery, (5) postsurgery baseline activity, (6) postsurgery shoaling behavior, and (7) postsurgery optomotor response. In the first phase, subjects were placed in the center area of the experimental tank. The subject swam alone and its trace was recorded for 20 min. Swimming distance was used as an index of activity. In this phase, subjects that stayed in one half of the tank for more than 70% of the time were eliminated. In the second phase, the basic procedure was the same as in the first, except that the stimulus fish was put into one of the side compartments 5 min after starting the session. Shoaling behavior was then recorded for 15 min. An area within 6 cm of the compartment where the stimulus fish was kept was defined as the shoaling area. Time spent within this area was used as an index of shoaling behavior. This phase was carried out on two successive days and the area where the stimulus fish was introduced was counterbalanced. The third phase was performed to examine visual and motor abilities. Subjects were required to swim five laps in each direction. Subjects that did not meet the criterion were eliminated. The subjects were then systematically divided into three groups to minimize group differences in activity and shoaling behavior. Each group was then given surgery:
Fig. 1. An example of brains from an intact fish (top), and a telencephalonablated fish (bottom). Scale bar=1 mm.
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
143
2.1.4. Surgery Subjects were anesthetized with 1% urethane solution and placed in a stereotaxic apparatus. Surgery was carried out under a microscope. The skull was drilled and removed to expose the brain. In the ablation group, the telencephalon was gently aspirated. In the OlT group, the olfactory tract was cut with scissors and forceps. In the sham group, the skull was removed without damaging the brain. The skulls of all subjects were then covered with gelform, resin, and dental cement. Fresh water, or 0.5% urethane solution, was circulated from a mouthpiece during surgery. Subjects were given 2 days to recover after surgery. 2.1.5. Histology At the end of the experiment, subjects were overanesthetized in 2% urethane solution and perfused with 0.6% saline and 10% formalin. Brains were removed and fixed with 10% formalin. Following fixation, brains in the ablation group were photographed, embedded in paraffin and sectioned at 5 Am with a microtome. The sections were stained with cresylecht violet and examined under a microscope. Fig. 2. A frontal section of the telencephalon-ablated brain at the level of the anterior commissure. The remaining preoptic area and optic chiasm are intact, whereas telencephalon has been removed.
the ablation group (n=8), the olfactory tract (OlT) group (n=6), and the sham group (n=6). The OlT group served as nonolfaction control, because telencephalic ablation damages olfactory input. After surgery, Phases 5 to 7 were conducted. In a previous study, using identical apparatus, we demonstrated that the number of stimulus fish (1, 2, or 5) did not affect the subject’s time spent near the stimulus [21]. Thus, 1 stimulus fish was used.
2.2. Results and discussion 2.2.1. Histology Fig. 1 shows the brains of an intact fish (top) and an ablated fish (bottom). All of the ablated fish had their telencephalon removed, giving a similar result to that seen on the bottom of Fig. 1. Most areas of the telencephalon were removed without damaging other brain areas. Histological examination showed that the telencephalon was completely removed, whereas the preoptic area and the optic chiasm remained intact in each fish (Fig. 2).
Fig. 3. Results of the activity phase in Experiment 1 (mean+S.D.). Statistical analysis was carried out by one-way analysis of variance (ANOVA) and Fisher’s PLSD. The ablation group differed significantly from the other groups (***P<.001).
144
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
2.2.2. Behavioral results 2.2.2.1. Optomotor response. One subject did not meet the criterion in the presurgery period and thus, was eliminated from further study. In the postsurgery period, all subjects met the required criterion. 2.2.2.2. Activity. Fig. 3 shows the results of the baseline activity phase. Surgery did not affect activity except in the case of the bilateral telencephalic ablation. One-way analysis of variance (ANOVA) was applied to analyze pre- and postsurgery data. No significant differences were seen presurgery [ F(2,17)=2.95, NS]; however, a significant effect was seen postsurgery [ F(2,17)=32.14, P<.0001]. Statistical analysis carried out by Fisher’s PLSD revealed that the ablation group differed significantly from the other groups ( P<.0001). 2.2.2.3. Shoaling behavior. Fig. 4 shows the results of the shoaling behavior phase. The vertical axis indicates the ratio of time spent within the shoaling area. Surgery did not affect the shoaling behavior of the fish, except for those in the ablation group. No significant differences were seen presurgery [ F(2,17)=1.16, NS], although a significant effect was seen postsurgery [ F(2,17)=6.13; P=.0099]. Statistical analysis using Fisher’s PLSD revealed that the ablation group differed from the other groups (vs. OlT; P=.0347, vs. sham; P=.0036). Telencephalic ablation reduced both the activity and the shoaling behavior of goldfish; however, because subjects showed an optomotor response, this reduction was not due to a simple visual or motor deficit. Moreover, the effects
were not due to loss of olfactory input caused by telencephalic ablation because the OlT group did not differ from the sham group.
3. Experiment 2 Experiment 1 demonstrated that telencephalic ablation decreased activity and shoaling behavior in goldfish. However, it is not clear whether these effects were caused by the lack of a specific part of the telencephalon. Experiment 2 examined the effects of partial lesions in the telencephalon on shoaling behavior. Lesions of the amygdala affect social behavior in mammals (for a review, see Ref. [1]). Thus, we hypothesized that social behavior in fish is controlled by a region of the brain homologous to the amygdala. It has been proposed that the dorsal part of the telencephalon, Dm, is homologous to the pallial amygdala of land vertebrates. Anatomically, the Dm lies between the ventral part of the telencephalon and the olfactory-tract-recipient Dp. Its topological position is similar to that of the pallial amygdala of frogs, which lies between the striatum and the olfactorytract-recipient lateral pallium [3]. Results of some behavioral studies support this proposal. For example, lesions in the Dm result in an increase in aggressive behavior, whereas more caudally located lesions result in a decrease in such behavior [4]. Electrical stimulation of the Dm results in changes in aggressive and reproductive behaviors [7], and lesions in the Dm impair performance of avoidance tasks, which probably relates to emotional memory [28]. Thus, the aim of Experiment 2 was to examine the
Fig. 4. Results of the shoaling behavior phase in Experiment 1 (mean+S.D.). The vertical axis indicates the ratio of time spent within 6 cm of the wall of the compartment where the stimulus fish was introduced. * and ** indicate significance levels P<.05 and .01, respectively.
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
145
Our hypothesis is based on two assumptions. Firstly, the Dm is homologous to the amygdala, and secondly, the area homologous to the amygdala in fish has similar functions to the amygdala of land vertebrates. Thus, if the Dm lesions impair shoaling behavior, there are two possible explanations. (1) The Dm is homologous to the amygdala of land vertebrates and has a similar function to the amygdala; or (2) the Dm is not homologous to the amygdala of land vertebrates but has a similar function to the amygdala. Alternatively, if the Dm lesions do not impair shoaling behavior, there are two possible explanations. (1) The Dm is not homologous to the amygdala of land vertebrates; or (2) the Dm is homologous but has a different function to that of the land vertebrates’ amygdala. 3.1. Methods 3.1.1. Animals Twelve goldfish were used as subjects. Their mean body length was 8.3 cm (range, 7.1 – 9.5 cm). The stimulus fish was identical to that used in Experiment 1.
Fig. 5. Representations of the lesions. Top: Dm lesion, bottom: Dl lesion.
role of one of the dorsal part of the telencephalon, the Dm, in shoaling behavior. Fish with lesions in the Dl region of the telencephalon were chosen as the control group because there is no evidence of Dl being homologous to the amygdala.
3.1.2. Procedure The basic procedure was the same as for Experiment 1. Subjects were assigned to two groups for surgery: the Dl group (n=6), in which the caudal part of the dorsolateral telencephalon was damaged; and the Dm group (n=6), in which the caudal part of the dorsomedial telencephalon was damaged. Damage was achieved by using a radiofrequency lesion generator (RFG-4A, Radionics) with a 0.25-mm diameter tip at 60 jC for 1 min. At the end of the experiment, the subject’s brains were perfused and fixed as per Experiment 1. Following fixation, brains were em-
Fig. 6. Results of the activity phase in Experiment 2 (+S.D.). The sham group in Experiment 1 is shown for comparison.
146
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
bedded in gelatin and sectioned at 50 Am with a microslicer. Sections were stained with cresylecht violet and examined under the microscope to reconstruct the damage.
and no effect was indicated [pre: F(2,15)=0.82, NS; post: F(2,15)=0.63, NS]. These results suggest that lesions in the Dl or Dm regions of the telencephalon are not responsible for the reduction of shoaling behavior seen in Experiment 1.
3.2. Results and discussion 3.2.1. Histology Fig. 5 illustrates the damage done to the telencephalon in the Dl and Dm groups in accordance with the brain atlas of goldfish [24]. Damage was seen from atlas locations +1.4 (level of anterior commissure) to +2.3. In both groups, additional damage was seen in the Dd region. Previous anatomical studies have shown that both Dl and Dm can be divided into subdivisions: ventral (Dlv) and dorsal (Dld) Dl, and ventral (Dmv) and dorsal (Dmd) Dm, respectively [19,20,34]. These subdivisions possibly have different functions. In the present study, only the dorsal part was damaged in both groups, although precise discrimination of these subdivisions with Nissl staining is difficult. 3.2.2. Behavioral results 3.2.2.1. Optomotor response. All subjects met the required criterion in the pre- and postsurgery period. Thus, visual and motor abilities were not damaged. 3.2.2.2. Activity and shoaling behavior. Fig. 6 shows the results of the activity phase. The results of the sham group in Experiment 1 are shown for comparison. One-way ANOVA was applied pre- and postsurgery, but no effect was indicated [pre: F(2,15)=3.33, NS; post: F(2,15)=1.44, NS]. Fig. 7 shows the results of the shoaling behavior phase. Each group showed a similar rate of shoaling. These data were also analyzed with the sham group from Experiment 1
4. General discussion The present study demonstrates that telencephalon-ablated fish have impaired activity and shoaling behavior, whereas Dm-, Dl-, and OlT-damaged fish do not. Impairment of shoaling behavior might be caused by sensory or motor deficits, although this possibility seems implausible because all subjects showed an optomotor response after surgery. Telencephalic ablation did not impair perception of the movement of stripes or the ability to swim. However, it has been suggested that the telencephalon has a role in the processing of visual stimuli [5]. Further study is needed to examine whether telencephalon-ablated fish can recognize more complex visual stimuli, such as conspecifics. In addition, the lack of olfaction is not responsible for the deficits in the present results because the OlT group did not differ from the sham group. Impairment of shoaling behavior might be caused by a reduction in activity, because ablated fish had both decreased activity and shoaling behavior. As shown in Fig. 8, data after surgery can be divided into two clusters, namely, the ablation group and the others. Separate analysis indicated a low correlation in the ablation group (r=.06) and in the other groups (r=.13). This suggests that telencephalic ablation reduced both activity and shoaling behavior without a correlation between them. Moreover, Figs. 3 and 8 show that the ablation group could swim. If decreased shoaling behavior was due to decreased activity, the ablation group could ap-
Fig. 7. Results of the shoaling behavior phase in Experiment 2 (+S.D.). The sham group in Experiment 1 is shown for comparison.
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
147
Fig. 8. Correlation between activity and shoaling behavior. Results can be divided into two clusters—the ablation group and the other groups.
proach the stimulus and stay there. However, they did not do so. Thus, it is plausible to assume that telencephalic ablation independently impaired activity and shoaling behavior. Several observations support the idea that impairment of shoaling behavior might be caused by a reduction in sensitivity to social stimuli. Telencephalon ablation impairs reproductive behavior [6,14,15]. In particular, lesions in the Vs – pVv impaired spawning behavior in both male and female goldfish [15]. This study used prostaglandin F2 alpha (PG) to induce normal female-type behavior in males. Although induced female-type behavior in males did not depend on olfactory input [32], Vs –pVv lesions reduced spawning behavior. These researchers claim that the impaired spawning behavior was due, not only to a lack of olfactory information, but also to insensitivity to social stimuli. In addition, they reported that lesions in the Vs – pVv regions reduced activity. These data are consistent with the present results, which suggest that telencephalon-ablated fish decreased their activity and shoaling behavior because of insensitivity to social stimuli. In Experiment 2, lesion sites were selected on the basis of anatomical studies. Our prediction was that, if the Dm region of the goldfish brain has similar functions to the mammalian amygdala, lesions in this region would alter shoaling behavior. However, lesions in the Dm or Dl regions did not affect activity or shoaling behavior. The results are unlikely to have resulted from a smaller damaged volume compared with the whole telencephalic ablation, because in another experiment in which lesions were made in both the Dm and Dl areas (in preparation), resulting in an area larger than that damaged in Experiment 2, shoaling behavior was also not affected. Although it is possible that the remaining ventral part of the Dm area has a function in shoaling behavior, the results of Experiment 2 suggest that Dm does not have function in shoaling behavior. Thus, the Dm area
of the goldfish brain is not homologous to amygdala of land vertebrates, or if it is homologous, it has a different function. The ventral parts of the telencephalon, Vs and Vc, could be homologous to the basal amygdala of amphibians, on the basis of topological position and absence of olfactory input. Vi, Vp, and NT form a transitional series from the olfactorytract-recipient Dp to the preoptic area, and all receive olfactory input. These parts of the telencephalon could be homologous to the pallial amygdala of amphibians [22]. Behaviorally, lesions in the Vs – pVv impair reproductive behavior, and a similarity of effects of lesions between Vs – pVv and the mammalian medial amygdala has been noted [14,15]. In contrast, the present study focused on nonsexual social behavior. In mammals, the amygdala mediates not only sexual but also nonsexual social behavior. For example, amygdaloid lesions in rats changed nonsexual social behavior, such as contact time and distance between individuals [9], and similar lesions impaired discrimination of conspecifics in mice [2]. The results in rats [9] are similar to those of the present study, except that activity increased in the amygdaloid-lesioned rats. If the Vs – pVv portions of the teleost telencephalon have functions similar to those of the mammalian amygdala, these portions may also mediate the nonsexual social behavior investigated in the present study. Although the present study did not directly examine the role of the ventral part of the goldfish telencephalon, the results suggest that the ventral part probably mediates shoaling behavior. Most studies of social behavior are carried out using mammals in which the entire amygdala is damaged [1], and the subdivisions are not considered. Thus, it is possible that portions of the goldfish brain homologous to both the subpallial and pallial amygdala in land vertebrates relate to shoaling behavior, and lesions in one of these portions has no effect. Further study with partial lesions will be necessary to investigate the relation-
148
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
ship between shoaling behavior and the telencephalon, and the homology between the teleost telencephalon and the amygdala of land vertebrates.
Acknowledgements This research was supported by the Fukuzawa Foundation (Keio University).
References [1] Bachevalier J. The amygdala, social behavior, and autism. In: Aggleton JP, editor. The amygdala. 2nd ed. Oxford: Oxford Univ. Press; 2000. p. 509 – 43. [2] Borlongan CV, Watanabe S. Failure to discriminate conspecifics in amygdaloid-lesioned mice. Pharmacol Biochem Behav 1994;48: 677 – 80. [3] Braford MR. Comparative aspects of forebrain organization in the ray-finned fishes: touchstones or not? Brain Behav Evol 1995;46: 259 – 74. [4] de Bruin JPC. Telencephalon and behavior in teleost fish. A neuroethological approach. In: Ebbesson SOE, editor. Comparative neurology of the telencephalon. New York: Plenum; 1980. p. 175 – 201. [5] Davis RE, Kassel J. Behavioral functions of the teleostean telencephalon. In: Northcutt RG, Davis RE, editors. Fish neurobiology, vol. 2. Ann Arbor: Univ. of Michigan Press; 1983. p. 237 – 63. [6] Davis RE, Kassel J, Schwagmeyer P. Telencephalic lesions and behavior in the teleost, Macropodus opercularis: reproduction, startle reaction, and operant behavior in the male. Behav Biol 1976;18: 165 – 77. [7] Demski LS. Behavioral effects of electrical stimulation of the brain. In: Northcutt RG, Davis RE, editors. Fish neurobiology, vol. 2. Ann Arbor: Univ. of Michigan Press; 1983. p. 317 – 59. [8] Flood NC, Overmier JB, Savage GE. Teleost telencephalon and learning: an interpretive review of data and hypotheses. Physiol Behav 1976;16:783 – 98. [9] Jonason KR, Enloe LJ. Alterations in social behavior following septal and amygdaloid lesions in the rat. J Comp Physiol Psychol 1971;75: 286 – 301. [10] Kassel J, Davis RE. Recovery of function following simultaneous and serial telencephalon ablation in the teleost, Macropodus opercularis. Behav Biol 1977;21:489 – 99. [11] Kassel J, Davis RE, Schwagmeyer P. Telencephalic lesions and behavior in the teleost, Macropodus opercularis: further analysis of reproductive and operant behavior in the male. Behav Biol 1976;18: 179 – 88. [12] Keenleyside MHA. Some aspects of the schooling behavior in fish. Behavior 1955;8:183 – 248. [13] Krause J, Godin JGJ. Shoal choice in banded killifish (Fundulus diapbanus, Teleostei, Cyprinodontidae): the effects of predation risk, fish size, species composition and size of shoals. Ethology 1994; 98:128 – 36. [14] Kyle AL, Peter RE. Effects of forebrain lesions on spawning behavior in the male goldfish. Physiol Behav 1982;28:1103 – 9. [15] Kyle AL, Stacey NE, Peter RE. Ventral telencephalic lesions: effects on bisexual behavior, activity, and olfaction in the male goldfish. Behav Neural Biol 1982;36:229 – 41.
[16] Laming PR, McKee M. Deficits in habituation of cardiac arousal responses incurred by telencephalic ablation in goldfish, Carassius auratus, and their relation to other telencephalic functions. J Comp Physiol Psychol 1981;95:460 – 7. [17] Magurran AE, Pitcher TJ. Foraging, timidity and shoal size in minnows and goldfish. Behav Ecol Sociobiol 1983;12:147 – 52. [18] McRobert SP, Bradner J. The influence of body coloration on shoaling preferences in fish. Anim Behav 1998;56:611 – 5. [19] Murakami T, Morita Y, Ito H. Extrinsic and intrinsic fiber connections of the telencephalon in a teleost, Sebastiscus marmoratus. J Comp Neurol 1983;216:115 – 31. [20] Nieuwenhuys R, Meek J. The telencephalon of actinopterygian fishes. In: Jones EG, Peters A, editors. Cerebral cortex, vol. 8A. New York: Plenum; 1990. p. 31 – 73. [21] Nishimura K, Yoshida M, Watanabe S. The effect on other individual presentations of the goldfish by FG7142 injection [in Japanese]. Jpn J Neuropsychopharmacol 2002;22:55 – 9. [22] Northcutt RG, Braford Jr MR. New observations on the organization and evolution of the telencephalon of actinopterygian fishes. In: Ebbesson SOE, editor. Comparative neurology of the telencephalon. New York: Plenum; 1980. p. 41 – 95. [23] Northcutt RG, Davis R. Telencephalic organization in ray-finned fishes. In: Northcutt RG, Davis RE, editors. Fish neurobiology, vol. 2. Ann Arbor: Univ. of Michigan Press; 1983. p. 203 – 36. [24] Peter RE, Gill VE. A steleotaxic atlas and technique for forebrain nuclei of the goldfish, Carassius auratus. J Comp Neurol 1975;159: 69 – 102. [25] Pitcher TJ. Heuristic definitions of fish shoaling behaviour. Anim Behav 1983;31:611 – 3. [26] Pitcher TJ, House AC. Foraging rules for group feeders, area copying depends upon food density in shoaling goldfish. Ethology 1987;76: 161 – 7. [27] Pitcher TJ, Magurran AE, Winfield IJ. Fish in larger shoals find food faster. Behav Ecol Sociobiol 1982;10:149 – 51. [28] Portavella M, Vargas JP, Torres B, Salas C. The effects of telencephalic pallial lesions on spatial, temporal, and emotional learning in goldfish. Brain Res Bull 2002;57:397 – 9. [29] Reebs SG, Saulnier N. The effect of hunger on shoal choice in golden shiners (Pisces: Cyprinidae, Notemigonus crysoleucas). Ethology 1997;103:642 – 52. [30] Salas C, Broglio C, Rodriguez F, Lopez JC, Portavella M, Torres B. Telencephalic ablation in goldfish impairs performance in a ‘spatial constancy’ problem but not in a cued one. Behav Brain Res 1996;79: 193 – 200. [31] Savege GE. Telencephalic lesions and avoidance behavior in the goldfish (Carassius auratus). Anim Behav 1969;17:362 – 73. [32] Stacey NE, Kyle AL. Effects of olfactory tract lesions on sexual and feeding behavior in the goldfish. Physiol Behav 1983;30: 621 – 8. [33] Street NE, Magurran AE, Pitcher TJ. The effects of increasing shoal size on handling time in goldfish, Carassius auratus L. J Fish Biol 1984;25:561 – 6. [34] Yamane Y, Yoshimoto M, Ito H. Area dorsalis pars lateralis of the telencephalon in a teleost (Sebastiscus marmoratus) can be divided into dorsal and ventral regions. Brain Behav Evol 1996;48: 338 – 49. [35] Yoshimoto M, Ito H. Structure and function of the telencephalon. [in Japanese]. In: Uematsu K, Oka Y, Ito H, editors. Neuroscience of fishes. Tokyo: Koseisha-koseikaku; 2002. p. 178 – 95.
Effects of telencephalic ablation on shoaling behavior in goldfish Kazutaka Shinozuka*, Shigeru Watanabe Department of Psychology, Keio University, Mita 2-15-45, Minato-ku, Tokyo, Japan Received 28 February 2003; received in revised form 10 October 2003; accepted 20 January 2004
Abstract The role of the telencephalon in the shoaling behavior of the goldfish, Carassius auratus, was investigated. Experiments were carried out in a tank divided into three compartments. Subjects were introduced into the center compartment of the tank. In the activity phase, subjects swam alone, and swimming distance was used as an index of activity. In the shoaling behavior phase, a stimulus fish was introduced into one of the side compartments, and the time spent near the side compartment by the subject was used as index of shoaling behavior. After these measurements were made, visual and motor abilities were examined using the optomotor response. Subjects then received surgery and the same procedure was repeated. In Experiment 1, the effects of total ablation of the telencephalon and a section of the olfactory tract (OlT) were examined. The ablation group exhibited reduced activity and shoaling behavior compared with the sham and OlT group. In Experiment 2, the role of the dorsal part of the telencephalon was examined after damaging the dorsomedial and dorsolateral telencephalon. Lesions in either portion had no effect and no simple visual or motor deficits were seen. These results suggest that the ventral part of the telencephalon mediates shoaling behavior. D 2004 Elsevier Inc. All rights reserved. Keywords: Goldfish; Shoaling behavior; Social behavior; Telencephalon
1. Introduction Numerous studies have shown that many factors affect shoaling behavior in fish, including number of members [12], body size [13], body coloration [18], hunger [29], and foraging behavior [26]. Shoaling has advantages (e.g., predator avoidance and food discovery) and disadvantages (e.g., competition for foraging). Fish tend to choose a shoal of fish with similar body size, because large fish in a shoal of small members may be striking and the risk of predation would be high. On the other hand, small fish in a shoal with large members may experience competition for food. Hungry fish tend to choose a small group rather than a large group because competition for food is reduced in the small group, although the risk of being preyed upon is increased [29]. Although many studies have investigated factors affecting shoaling behavior, the region of the brain involved in shoaling behavior is not known. The teleost telencephalon, which was once considered an olfactory system, is now known to be involved in more complex functions, such as * Corresponding author. Tel.: +81-354-433896; fax: +81-354-433897. E-mail address: [email protected] (K. Shinozuka). 0031-9384/$ – see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2004.01.005
reproductive behavior and learning. For example, telencephalon-ablated fish show reduced spawning behavior [6,10, 11]. Telencephalic ablation also impairs some types of learning, including habituation [16], complex operant tasks [8], avoidance tasks [31], and spatial learning [30]. Anatomically, the teleost telencephalon can be divided into two parts—the area dorsalis telencephali (D) and the area ventralis telencephali (V). Although differences exist between species [23], in general, the dorsal part consists of five substructures: the pars medialis (Dm), the pars dorsalis (Dd), the pars lateralis (Dl), the pars centralis (Dc), and the pars posterior (Dp). The ventral part consists of seven substructures: the pars ventralis (Vv), the pars dorsalis (Vd), the pars lateralis (Vl), the pars supracommissuralis (Vs), the pars postocommissuralis (Vp), the pars intermedia (Vi), and the nucleus entopeduncularis (E) [35]. Homology of these substructures to the brains of land vertebrates cannot be easily determined because the teleost brain undergoes a different process of development, called eversion. As a result of eversion, the pallium of the teleost telencephalon reverses mediolaterally compared with land vertebrates [35]. On the basis of neuronal connections, morphology, and histochemistry, the substructures Vs and area ventralis telencephali pars centralis (Vc) are considered homologous to
142
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
the basal amygdala of land vertebrates, and Vp, Vi, and nucleus taenia (NT) are considered homologous to the pallial amygdala [22]. More recently, however, it was proposed that Dm is homologous to the pallial amygdala [3]. Thus, it is difficult to draw homologies between the parts of the teleost telencephalon and their counterparts in other vertebrates. Using behavioral studies, some researchers claim there is homology between the teleost telencephalon and the mammalian limbic system [8,15,28]. Although most of these studies examined the effects of total ablation of the telencephalon, a few investigated the effects of partial lesions. For example, Dl lesions impaired performance in spatial learning tasks, which suggests that Dl is homologous to the hippocampus of land vertebrates. Dm lesions impaired performance on the shuttle avoidance task, indicating that the effect was caused by disruption of emotional memory, which suggests that Dm could be homologous to the pallial amygdala of land vertebrates [28]. Lesions in the Vs and posterior part of Vv (pVv) regions blocked the initiation of spawning behavior, suggesting homology between these regions and the medial amygdala of land vertebrates [14]. The purpose of the present study was to investigate whether the telencephalon mediates shoaling behavior in goldfish. Goldfish have been extensively used for behavioral studies, and they do shoal [17,25,27,33]. The effects of ablation of the total telencephalon were examined in Experiment 1, and the effects of partial lesions in the dorsal part of the telencephalon were examined in Experiment 2.
2. Experiment 1 2.1. Methods 2.1.1. Animals Twenty-one goldfish (Carassius auratus) were obtained from a local dealer. Their mean body length was 8.0 cm (range, 7.4– 9.0 cm). One randomly selected fish was used as a stimulus fish and the rest were used as subjects. The fish were housed in a Plexiglas tank of filtered water. Experiments were not begun until at least 7 days after the fish arrived. The fish were fed daily on commercial goldfish food. 2.1.2. Apparatus The tank in which experiments were carried out was 60 cm wide, 30 cm long and 36 cm deep, and was divided into three areas by transparent glass panels. The center area was 30 cm wide and 30 cm long. The side areas were 15 cm wide and 30 cm long. The water depth was 5 cm. A CCD camera (XC-711, SONY) set above the tank sent its signal to the Chromascan (OKK) video tracker (G220, OKK) that measured the position of the goldfish by color detection. An IBM PC/AT compatible computer recorded the data.
To examine the optomotor response, a circular tank (32 cm diameter and 20 cm deep) was used. A glass cylinder (5.5 cm diameter) was set in the center of the tank. The depth of the water in the tank was 5 cm. A drum (34 cm diameter and 21 cm high) painted with vertical black and white lines (3.2 cm wide) was placed around the tank and rotated clockwise or counterclockwise 15 times/min. During this task, fish tend to chase the rotation of the drum, allowing visual and motor ability to be measured by their optomotor response. 2.1.3. Procedure The experiment consisted of seven phases: (1) presurgery baseline activity, (2) presurgery shoaling behavior, (3) presurgery optomotor response, (4) surgery, (5) postsurgery baseline activity, (6) postsurgery shoaling behavior, and (7) postsurgery optomotor response. In the first phase, subjects were placed in the center area of the experimental tank. The subject swam alone and its trace was recorded for 20 min. Swimming distance was used as an index of activity. In this phase, subjects that stayed in one half of the tank for more than 70% of the time were eliminated. In the second phase, the basic procedure was the same as in the first, except that the stimulus fish was put into one of the side compartments 5 min after starting the session. Shoaling behavior was then recorded for 15 min. An area within 6 cm of the compartment where the stimulus fish was kept was defined as the shoaling area. Time spent within this area was used as an index of shoaling behavior. This phase was carried out on two successive days and the area where the stimulus fish was introduced was counterbalanced. The third phase was performed to examine visual and motor abilities. Subjects were required to swim five laps in each direction. Subjects that did not meet the criterion were eliminated. The subjects were then systematically divided into three groups to minimize group differences in activity and shoaling behavior. Each group was then given surgery:
Fig. 1. An example of brains from an intact fish (top), and a telencephalonablated fish (bottom). Scale bar=1 mm.
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
143
2.1.4. Surgery Subjects were anesthetized with 1% urethane solution and placed in a stereotaxic apparatus. Surgery was carried out under a microscope. The skull was drilled and removed to expose the brain. In the ablation group, the telencephalon was gently aspirated. In the OlT group, the olfactory tract was cut with scissors and forceps. In the sham group, the skull was removed without damaging the brain. The skulls of all subjects were then covered with gelform, resin, and dental cement. Fresh water, or 0.5% urethane solution, was circulated from a mouthpiece during surgery. Subjects were given 2 days to recover after surgery. 2.1.5. Histology At the end of the experiment, subjects were overanesthetized in 2% urethane solution and perfused with 0.6% saline and 10% formalin. Brains were removed and fixed with 10% formalin. Following fixation, brains in the ablation group were photographed, embedded in paraffin and sectioned at 5 Am with a microtome. The sections were stained with cresylecht violet and examined under a microscope. Fig. 2. A frontal section of the telencephalon-ablated brain at the level of the anterior commissure. The remaining preoptic area and optic chiasm are intact, whereas telencephalon has been removed.
the ablation group (n=8), the olfactory tract (OlT) group (n=6), and the sham group (n=6). The OlT group served as nonolfaction control, because telencephalic ablation damages olfactory input. After surgery, Phases 5 to 7 were conducted. In a previous study, using identical apparatus, we demonstrated that the number of stimulus fish (1, 2, or 5) did not affect the subject’s time spent near the stimulus [21]. Thus, 1 stimulus fish was used.
2.2. Results and discussion 2.2.1. Histology Fig. 1 shows the brains of an intact fish (top) and an ablated fish (bottom). All of the ablated fish had their telencephalon removed, giving a similar result to that seen on the bottom of Fig. 1. Most areas of the telencephalon were removed without damaging other brain areas. Histological examination showed that the telencephalon was completely removed, whereas the preoptic area and the optic chiasm remained intact in each fish (Fig. 2).
Fig. 3. Results of the activity phase in Experiment 1 (mean+S.D.). Statistical analysis was carried out by one-way analysis of variance (ANOVA) and Fisher’s PLSD. The ablation group differed significantly from the other groups (***P<.001).
144
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
2.2.2. Behavioral results 2.2.2.1. Optomotor response. One subject did not meet the criterion in the presurgery period and thus, was eliminated from further study. In the postsurgery period, all subjects met the required criterion. 2.2.2.2. Activity. Fig. 3 shows the results of the baseline activity phase. Surgery did not affect activity except in the case of the bilateral telencephalic ablation. One-way analysis of variance (ANOVA) was applied to analyze pre- and postsurgery data. No significant differences were seen presurgery [ F(2,17)=2.95, NS]; however, a significant effect was seen postsurgery [ F(2,17)=32.14, P<.0001]. Statistical analysis carried out by Fisher’s PLSD revealed that the ablation group differed significantly from the other groups ( P<.0001). 2.2.2.3. Shoaling behavior. Fig. 4 shows the results of the shoaling behavior phase. The vertical axis indicates the ratio of time spent within the shoaling area. Surgery did not affect the shoaling behavior of the fish, except for those in the ablation group. No significant differences were seen presurgery [ F(2,17)=1.16, NS], although a significant effect was seen postsurgery [ F(2,17)=6.13; P=.0099]. Statistical analysis using Fisher’s PLSD revealed that the ablation group differed from the other groups (vs. OlT; P=.0347, vs. sham; P=.0036). Telencephalic ablation reduced both the activity and the shoaling behavior of goldfish; however, because subjects showed an optomotor response, this reduction was not due to a simple visual or motor deficit. Moreover, the effects
were not due to loss of olfactory input caused by telencephalic ablation because the OlT group did not differ from the sham group.
3. Experiment 2 Experiment 1 demonstrated that telencephalic ablation decreased activity and shoaling behavior in goldfish. However, it is not clear whether these effects were caused by the lack of a specific part of the telencephalon. Experiment 2 examined the effects of partial lesions in the telencephalon on shoaling behavior. Lesions of the amygdala affect social behavior in mammals (for a review, see Ref. [1]). Thus, we hypothesized that social behavior in fish is controlled by a region of the brain homologous to the amygdala. It has been proposed that the dorsal part of the telencephalon, Dm, is homologous to the pallial amygdala of land vertebrates. Anatomically, the Dm lies between the ventral part of the telencephalon and the olfactory-tract-recipient Dp. Its topological position is similar to that of the pallial amygdala of frogs, which lies between the striatum and the olfactorytract-recipient lateral pallium [3]. Results of some behavioral studies support this proposal. For example, lesions in the Dm result in an increase in aggressive behavior, whereas more caudally located lesions result in a decrease in such behavior [4]. Electrical stimulation of the Dm results in changes in aggressive and reproductive behaviors [7], and lesions in the Dm impair performance of avoidance tasks, which probably relates to emotional memory [28]. Thus, the aim of Experiment 2 was to examine the
Fig. 4. Results of the shoaling behavior phase in Experiment 1 (mean+S.D.). The vertical axis indicates the ratio of time spent within 6 cm of the wall of the compartment where the stimulus fish was introduced. * and ** indicate significance levels P<.05 and .01, respectively.
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
145
Our hypothesis is based on two assumptions. Firstly, the Dm is homologous to the amygdala, and secondly, the area homologous to the amygdala in fish has similar functions to the amygdala of land vertebrates. Thus, if the Dm lesions impair shoaling behavior, there are two possible explanations. (1) The Dm is homologous to the amygdala of land vertebrates and has a similar function to the amygdala; or (2) the Dm is not homologous to the amygdala of land vertebrates but has a similar function to the amygdala. Alternatively, if the Dm lesions do not impair shoaling behavior, there are two possible explanations. (1) The Dm is not homologous to the amygdala of land vertebrates; or (2) the Dm is homologous but has a different function to that of the land vertebrates’ amygdala. 3.1. Methods 3.1.1. Animals Twelve goldfish were used as subjects. Their mean body length was 8.3 cm (range, 7.1 – 9.5 cm). The stimulus fish was identical to that used in Experiment 1.
Fig. 5. Representations of the lesions. Top: Dm lesion, bottom: Dl lesion.
role of one of the dorsal part of the telencephalon, the Dm, in shoaling behavior. Fish with lesions in the Dl region of the telencephalon were chosen as the control group because there is no evidence of Dl being homologous to the amygdala.
3.1.2. Procedure The basic procedure was the same as for Experiment 1. Subjects were assigned to two groups for surgery: the Dl group (n=6), in which the caudal part of the dorsolateral telencephalon was damaged; and the Dm group (n=6), in which the caudal part of the dorsomedial telencephalon was damaged. Damage was achieved by using a radiofrequency lesion generator (RFG-4A, Radionics) with a 0.25-mm diameter tip at 60 jC for 1 min. At the end of the experiment, the subject’s brains were perfused and fixed as per Experiment 1. Following fixation, brains were em-
Fig. 6. Results of the activity phase in Experiment 2 (+S.D.). The sham group in Experiment 1 is shown for comparison.
146
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
bedded in gelatin and sectioned at 50 Am with a microslicer. Sections were stained with cresylecht violet and examined under the microscope to reconstruct the damage.
and no effect was indicated [pre: F(2,15)=0.82, NS; post: F(2,15)=0.63, NS]. These results suggest that lesions in the Dl or Dm regions of the telencephalon are not responsible for the reduction of shoaling behavior seen in Experiment 1.
3.2. Results and discussion 3.2.1. Histology Fig. 5 illustrates the damage done to the telencephalon in the Dl and Dm groups in accordance with the brain atlas of goldfish [24]. Damage was seen from atlas locations +1.4 (level of anterior commissure) to +2.3. In both groups, additional damage was seen in the Dd region. Previous anatomical studies have shown that both Dl and Dm can be divided into subdivisions: ventral (Dlv) and dorsal (Dld) Dl, and ventral (Dmv) and dorsal (Dmd) Dm, respectively [19,20,34]. These subdivisions possibly have different functions. In the present study, only the dorsal part was damaged in both groups, although precise discrimination of these subdivisions with Nissl staining is difficult. 3.2.2. Behavioral results 3.2.2.1. Optomotor response. All subjects met the required criterion in the pre- and postsurgery period. Thus, visual and motor abilities were not damaged. 3.2.2.2. Activity and shoaling behavior. Fig. 6 shows the results of the activity phase. The results of the sham group in Experiment 1 are shown for comparison. One-way ANOVA was applied pre- and postsurgery, but no effect was indicated [pre: F(2,15)=3.33, NS; post: F(2,15)=1.44, NS]. Fig. 7 shows the results of the shoaling behavior phase. Each group showed a similar rate of shoaling. These data were also analyzed with the sham group from Experiment 1
4. General discussion The present study demonstrates that telencephalon-ablated fish have impaired activity and shoaling behavior, whereas Dm-, Dl-, and OlT-damaged fish do not. Impairment of shoaling behavior might be caused by sensory or motor deficits, although this possibility seems implausible because all subjects showed an optomotor response after surgery. Telencephalic ablation did not impair perception of the movement of stripes or the ability to swim. However, it has been suggested that the telencephalon has a role in the processing of visual stimuli [5]. Further study is needed to examine whether telencephalon-ablated fish can recognize more complex visual stimuli, such as conspecifics. In addition, the lack of olfaction is not responsible for the deficits in the present results because the OlT group did not differ from the sham group. Impairment of shoaling behavior might be caused by a reduction in activity, because ablated fish had both decreased activity and shoaling behavior. As shown in Fig. 8, data after surgery can be divided into two clusters, namely, the ablation group and the others. Separate analysis indicated a low correlation in the ablation group (r=.06) and in the other groups (r=.13). This suggests that telencephalic ablation reduced both activity and shoaling behavior without a correlation between them. Moreover, Figs. 3 and 8 show that the ablation group could swim. If decreased shoaling behavior was due to decreased activity, the ablation group could ap-
Fig. 7. Results of the shoaling behavior phase in Experiment 2 (+S.D.). The sham group in Experiment 1 is shown for comparison.
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
147
Fig. 8. Correlation between activity and shoaling behavior. Results can be divided into two clusters—the ablation group and the other groups.
proach the stimulus and stay there. However, they did not do so. Thus, it is plausible to assume that telencephalic ablation independently impaired activity and shoaling behavior. Several observations support the idea that impairment of shoaling behavior might be caused by a reduction in sensitivity to social stimuli. Telencephalon ablation impairs reproductive behavior [6,14,15]. In particular, lesions in the Vs – pVv impaired spawning behavior in both male and female goldfish [15]. This study used prostaglandin F2 alpha (PG) to induce normal female-type behavior in males. Although induced female-type behavior in males did not depend on olfactory input [32], Vs –pVv lesions reduced spawning behavior. These researchers claim that the impaired spawning behavior was due, not only to a lack of olfactory information, but also to insensitivity to social stimuli. In addition, they reported that lesions in the Vs – pVv regions reduced activity. These data are consistent with the present results, which suggest that telencephalon-ablated fish decreased their activity and shoaling behavior because of insensitivity to social stimuli. In Experiment 2, lesion sites were selected on the basis of anatomical studies. Our prediction was that, if the Dm region of the goldfish brain has similar functions to the mammalian amygdala, lesions in this region would alter shoaling behavior. However, lesions in the Dm or Dl regions did not affect activity or shoaling behavior. The results are unlikely to have resulted from a smaller damaged volume compared with the whole telencephalic ablation, because in another experiment in which lesions were made in both the Dm and Dl areas (in preparation), resulting in an area larger than that damaged in Experiment 2, shoaling behavior was also not affected. Although it is possible that the remaining ventral part of the Dm area has a function in shoaling behavior, the results of Experiment 2 suggest that Dm does not have function in shoaling behavior. Thus, the Dm area
of the goldfish brain is not homologous to amygdala of land vertebrates, or if it is homologous, it has a different function. The ventral parts of the telencephalon, Vs and Vc, could be homologous to the basal amygdala of amphibians, on the basis of topological position and absence of olfactory input. Vi, Vp, and NT form a transitional series from the olfactorytract-recipient Dp to the preoptic area, and all receive olfactory input. These parts of the telencephalon could be homologous to the pallial amygdala of amphibians [22]. Behaviorally, lesions in the Vs – pVv impair reproductive behavior, and a similarity of effects of lesions between Vs – pVv and the mammalian medial amygdala has been noted [14,15]. In contrast, the present study focused on nonsexual social behavior. In mammals, the amygdala mediates not only sexual but also nonsexual social behavior. For example, amygdaloid lesions in rats changed nonsexual social behavior, such as contact time and distance between individuals [9], and similar lesions impaired discrimination of conspecifics in mice [2]. The results in rats [9] are similar to those of the present study, except that activity increased in the amygdaloid-lesioned rats. If the Vs – pVv portions of the teleost telencephalon have functions similar to those of the mammalian amygdala, these portions may also mediate the nonsexual social behavior investigated in the present study. Although the present study did not directly examine the role of the ventral part of the goldfish telencephalon, the results suggest that the ventral part probably mediates shoaling behavior. Most studies of social behavior are carried out using mammals in which the entire amygdala is damaged [1], and the subdivisions are not considered. Thus, it is possible that portions of the goldfish brain homologous to both the subpallial and pallial amygdala in land vertebrates relate to shoaling behavior, and lesions in one of these portions has no effect. Further study with partial lesions will be necessary to investigate the relation-
148
K. Shinozuka, S. Watanabe / Physiology & Behavior 81 (2004) 141–148
ship between shoaling behavior and the telencephalon, and the homology between the teleost telencephalon and the amygdala of land vertebrates.
Acknowledgements This research was supported by the Fukuzawa Foundation (Keio University).
References [1] Bachevalier J. The amygdala, social behavior, and autism. In: Aggleton JP, editor. The amygdala. 2nd ed. Oxford: Oxford Univ. Press; 2000. p. 509 – 43. [2] Borlongan CV, Watanabe S. Failure to discriminate conspecifics in amygdaloid-lesioned mice. Pharmacol Biochem Behav 1994;48: 677 – 80. [3] Braford MR. Comparative aspects of forebrain organization in the ray-finned fishes: touchstones or not? Brain Behav Evol 1995;46: 259 – 74. [4] de Bruin JPC. Telencephalon and behavior in teleost fish. A neuroethological approach. In: Ebbesson SOE, editor. Comparative neurology of the telencephalon. New York: Plenum; 1980. p. 175 – 201. [5] Davis RE, Kassel J. Behavioral functions of the teleostean telencephalon. In: Northcutt RG, Davis RE, editors. Fish neurobiology, vol. 2. Ann Arbor: Univ. of Michigan Press; 1983. p. 237 – 63. [6] Davis RE, Kassel J, Schwagmeyer P. Telencephalic lesions and behavior in the teleost, Macropodus opercularis: reproduction, startle reaction, and operant behavior in the male. Behav Biol 1976;18: 165 – 77. [7] Demski LS. Behavioral effects of electrical stimulation of the brain. In: Northcutt RG, Davis RE, editors. Fish neurobiology, vol. 2. Ann Arbor: Univ. of Michigan Press; 1983. p. 317 – 59. [8] Flood NC, Overmier JB, Savage GE. Teleost telencephalon and learning: an interpretive review of data and hypotheses. Physiol Behav 1976;16:783 – 98. [9] Jonason KR, Enloe LJ. Alterations in social behavior following septal and amygdaloid lesions in the rat. J Comp Physiol Psychol 1971;75: 286 – 301. [10] Kassel J, Davis RE. Recovery of function following simultaneous and serial telencephalon ablation in the teleost, Macropodus opercularis. Behav Biol 1977;21:489 – 99. [11] Kassel J, Davis RE, Schwagmeyer P. Telencephalic lesions and behavior in the teleost, Macropodus opercularis: further analysis of reproductive and operant behavior in the male. Behav Biol 1976;18: 179 – 88. [12] Keenleyside MHA. Some aspects of the schooling behavior in fish. Behavior 1955;8:183 – 248. [13] Krause J, Godin JGJ. Shoal choice in banded killifish (Fundulus diapbanus, Teleostei, Cyprinodontidae): the effects of predation risk, fish size, species composition and size of shoals. Ethology 1994; 98:128 – 36. [14] Kyle AL, Peter RE. Effects of forebrain lesions on spawning behavior in the male goldfish. Physiol Behav 1982;28:1103 – 9. [15] Kyle AL, Stacey NE, Peter RE. Ventral telencephalic lesions: effects on bisexual behavior, activity, and olfaction in the male goldfish. Behav Neural Biol 1982;36:229 – 41.
[16] Laming PR, McKee M. Deficits in habituation of cardiac arousal responses incurred by telencephalic ablation in goldfish, Carassius auratus, and their relation to other telencephalic functions. J Comp Physiol Psychol 1981;95:460 – 7. [17] Magurran AE, Pitcher TJ. Foraging, timidity and shoal size in minnows and goldfish. Behav Ecol Sociobiol 1983;12:147 – 52. [18] McRobert SP, Bradner J. The influence of body coloration on shoaling preferences in fish. Anim Behav 1998;56:611 – 5. [19] Murakami T, Morita Y, Ito H. Extrinsic and intrinsic fiber connections of the telencephalon in a teleost, Sebastiscus marmoratus. J Comp Neurol 1983;216:115 – 31. [20] Nieuwenhuys R, Meek J. The telencephalon of actinopterygian fishes. In: Jones EG, Peters A, editors. Cerebral cortex, vol. 8A. New York: Plenum; 1990. p. 31 – 73. [21] Nishimura K, Yoshida M, Watanabe S. The effect on other individual presentations of the goldfish by FG7142 injection [in Japanese]. Jpn J Neuropsychopharmacol 2002;22:55 – 9. [22] Northcutt RG, Braford Jr MR. New observations on the organization and evolution of the telencephalon of actinopterygian fishes. In: Ebbesson SOE, editor. Comparative neurology of the telencephalon. New York: Plenum; 1980. p. 41 – 95. [23] Northcutt RG, Davis R. Telencephalic organization in ray-finned fishes. In: Northcutt RG, Davis RE, editors. Fish neurobiology, vol. 2. Ann Arbor: Univ. of Michigan Press; 1983. p. 203 – 36. [24] Peter RE, Gill VE. A steleotaxic atlas and technique for forebrain nuclei of the goldfish, Carassius auratus. J Comp Neurol 1975;159: 69 – 102. [25] Pitcher TJ. Heuristic definitions of fish shoaling behaviour. Anim Behav 1983;31:611 – 3. [26] Pitcher TJ, House AC. Foraging rules for group feeders, area copying depends upon food density in shoaling goldfish. Ethology 1987;76: 161 – 7. [27] Pitcher TJ, Magurran AE, Winfield IJ. Fish in larger shoals find food faster. Behav Ecol Sociobiol 1982;10:149 – 51. [28] Portavella M, Vargas JP, Torres B, Salas C. The effects of telencephalic pallial lesions on spatial, temporal, and emotional learning in goldfish. Brain Res Bull 2002;57:397 – 9. [29] Reebs SG, Saulnier N. The effect of hunger on shoal choice in golden shiners (Pisces: Cyprinidae, Notemigonus crysoleucas). Ethology 1997;103:642 – 52. [30] Salas C, Broglio C, Rodriguez F, Lopez JC, Portavella M, Torres B. Telencephalic ablation in goldfish impairs performance in a ‘spatial constancy’ problem but not in a cued one. Behav Brain Res 1996;79: 193 – 200. [31] Savege GE. Telencephalic lesions and avoidance behavior in the goldfish (Carassius auratus). Anim Behav 1969;17:362 – 73. [32] Stacey NE, Kyle AL. Effects of olfactory tract lesions on sexual and feeding behavior in the goldfish. Physiol Behav 1983;30: 621 – 8. [33] Street NE, Magurran AE, Pitcher TJ. The effects of increasing shoal size on handling time in goldfish, Carassius auratus L. J Fish Biol 1984;25:561 – 6. [34] Yamane Y, Yoshimoto M, Ito H. Area dorsalis pars lateralis of the telencephalon in a teleost (Sebastiscus marmoratus) can be divided into dorsal and ventral regions. Brain Behav Evol 1996;48: 338 – 49. [35] Yoshimoto M, Ito H. Structure and function of the telencephalon. [in Japanese]. In: Uematsu K, Oka Y, Ito H, editors. Neuroscience of fishes. Tokyo: Koseisha-koseikaku; 2002. p. 178 – 95.