- Email: [email protected]
Effects of telencephalic ablation on habituation of arousal responses, within and between daily training sessions in goldfish
BEHAVIORAL AND NEURAL BIOLOGY
49, 83-96 (1988)
Effects of Telencephalic Ablation on Habituation of Arousal Responses, within and between Daily Training Sessions in Goldfish D. J. ROONEY AND P. R. LAMING Department of Zoology, The Queen's University of Belfast, Belfast BT7 1NN, United Kingdom Goldfish Carassius auratus were presented with a moving shadow stimulus at 2-min intervals and their cardiac and ventilatory responses were monitored. Normal fish, fish with their telencephalon ablated, and those with sham operations were compared for responsiveness and habituation to repeatedly presented stimuli over the 3-day test period. While all groups showed increased habituation on successive days testing, fish with their telencephalon ablated showed significantly slower response habituation within the daily test sessions when compared with the control groups. Subjects with their telencephalon ablated also showed a tendency for increased responsiveness on initial stimulus presentation and poorer retention between days of information relating to the eliciting test stimulus. © 1988 Academic Press, Inc.
Ablation of the teleostean telencephalon has been shown to impair habituation of nonspecific arousal responses in a number of studies (Laming & McKee, 1981; Laming & Ennis, 1982). Localized lesions in the posterior dorsomedial (Laming & Hornby, 1981) or dorsocentral telencephalic regions (Rooney & Laming, 1986a) have also resulted in significant deficits in habituation of the arousal response of fish to tap or visual stimuli. Extensive damage to the dorsolateral telencephalic areas of the squirrel fish Holocentrus rufus had no effect on habituation of arousal response to visual stimuli though posterior-medial telencephalic damage severely impaired habituation (Laming, 1987). A recent study by Marino-Neto and Sabbatini (1983) reports that lesions to the posterior area dorsalis pars medialis (Dmp) and commissura telencephali dorsalis (CTD) regions of the telencephalon in Siamese fighting fish (Betta splendens) accelerate habituation of arousal responses to tap stimuli: the result opposite to that obtained in Rutilus by Laming & Hornby (1981). Dmp/CTD lesions also resulted in hyperreactivity to novel environmental stimuli in Betta splendens, a finding not demonstrated 83 0163-1047/88 $3.00 Copyright © 1988 by Academic Press, Inc All rights of reproduction in any form reserved.
84
ROONEY AND LAMING
following telencephalic ablation or lesions in Carassius (Laming & McKee, 1981; Laming & Hornby, 1981). The differing findings may be due to species differences, as the CTD is an uncommon structure in actinopterygian fish (Nieuwenhuys, 1963) and its fine connections are unknown. It is also difficult to make close comparisons between Marino-Neto and Sabbatini's (1983) study and earlier work as in their study they investigated combined startle and arousal responses. Habituation of the former is unaffected by telencephalic ablation (Laming & Ennis, 1982). Another reported deficit following ablation of the teleostean telencephalon in species-specific behavioral studies has been in the consolidation of learning. Peeke et al.'s (1972) investigation into the effects of telencephalic ablation on habituation of predatory behavior in Carassius reported no short-term (within session) habituation deficit but fish with their telencephalon ablated showed no savings between days of habituation. Peeke et al. (1972) interpreted these results as evidence of impaired long-term memory but an intact short-term memory following telencephalic ablation. Similar disruption of long-term memory, resulting in deficits in foraging behavior after telencephalic ablation, in Carassius have also been found by Peeke and Gordon (1981). Gordon (1979) also reports disruption of memory following telencephalic ablation in a taste aversion study. Goldfish with their telencephalon ablated were unable to retain a taste-illness association from one trial to the next to avoid illness associated with LiC1 injection. Increased habituation on repeated testing is one of the parametric characteristics of habituation (Thompson & Spencer, 1966). Thus, an investigation into the effects of telencephalic ablation on habituation of the physiological correlates of the behavioral arousal response during successive days testing should clarify telencephalic influences on shortterm and long-term habituation. It should also address both the arousal and memory hypotheses for telencephalic function. This was the rationale for the present experiment in which the habituation of cardiac and ventilatory arousal responses to a moving shadow stimulus in Carassius were studied. MATERIALS AND METHODS Subjects
Goldfish, Carassius auratus, were obtained from a local supplier. They were maintained in 2 × 0.5 × 0.5-m glass aquaria of aerated filtered water at 12 _+ 2°C for 4 weeks prior to experiments. These took place between April and early June to avoid any seasonal variation in responsiveness (Laming & Savage, 1981). Thirty-four animals, 9-12 cm in length were randomly selected from stock tanks and assigned to one of three groups with differing degrees of surgical treatment. These were (1)
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
85
fish with the telencephalic lobes exposed and forebrain ablated by aspiration (A); (2) fish with the forebrain exposed and the telencephalic lobes left intact, the sham operated (S) group; and (3) fish with no cranial surgery, the normal (N) group.
ECG and Ventilatory Recording Procedures The ECG and ventilatory recording procedures were as described in Laming and Savage (1980). Briefly, fish were anesthetized by immersion in a solution of 1 : 10,000 MS222 (tricaine methane sulphonate) until opercular movements ceased. All fish were implanted with electrocardiogram (ECG) electrodes using the method of Roberts et al. (1973). A ventilatory catheter connecting the buccal cavity to a pressure transducer was also implanted. ECGs and ventilations were monitored by a Beckman R411 dynograph.
Cranial Surgery (I) Ablations (A). The 13 fish randomly selected for forebrain ablations were placed in the fish clamp, and their gills were perfused with a 1 : 20,000 solution of MS222 via a tube to the mouth. A 3-mm diameter observation hole was opened in the cranium midway between the eyes. Aspiration of the fatty cerebrospinal fluid exposed the telencephalic lobes which were then ablated by aspiration. Following telencephalic ablation the cranial cavity was filled with teleost saline and the inspection hole was sealed with a sterile cork and Eastman 910 adhesive (Laming, 1980). (2) Sham operation (S). The 10 sham operated fish underwent identical surgery though the telencephalic lobes were left intact. (3) " N o r m a l " fish (N) had no cranial surgery and acted as controls for the sham operation. All fish were revived by perfusion of their gills with fresh water, and allowed to recover in stock aquaria for 48 h prior to testing.
The Experimental Situation and Stimuli After recovery, fish were placed in a 22 x 6 x 10-cm-high wire trough placed centrally in a 35 x 25 x 25-cm-high perspex aquarium containing 12 cm of aerated water at 12 _+ 2°C. The aquarium was mounted on sponge rubber and situated in a 46 x 36 x 26-cm-high black box. The lid of the box contained the apparatus for providing the moving shadow and a longitudinally placed 60 W tungsten filament strip light gave a background illumination of 90 lux. A black screen (35 x 25 cm) fitted below the light incorporated a 10 x 5-cm central aperture. A rotating 22-cm diameter clear Perspex disk, split into six segments, alternatively clear and dark, passed between the light source and screen aperture at a speed of 0.7 cm/s. The stimulus was automatically presented for 10 s in every 120 s, allowing one dark segment to pass the aperture during
86
ROONEY AND LAMING
a stimulus presentation, starting and ending in a clear field. The precise position at which the stimulus ended was controlled by a photocell. The moving shadow approached the fish's visual field from behind as this has been shown to consistently elicit arousal responses (Laming & Savage, 1980). The fish were left for 1 h in the test tank to adapt to the situation prior to the presentation of stimuli. At the end of this period, spontaneous cardiac and ventilatory rate changes were monitored during two consecutive 10-s periods, to provide a baseline for responses to the stimuli which were then serially presented. Habituation of a response in an individual fish was considered to have occurred on the second of two consecutive stimulus presentations in which the ratio of the longest interbeat/interbreath interval in the 10-s stimulus period compared with that in the 10-s prestimulus period fell within the 95% confidence limits of the control (no stimulus) variation from unity (Laming & McKee, 1981). Following at least 20 stimulus presentations, fish were given a tap stimulus for one trial to investigate response dishabituation. The tap was delivered to the side of the box by a Perspex rod mounted on a remotely operated 24 V dc solenoid. Following presentation of the tap, resumption of stimulation with the moving shadow continued for five more trials. The fish were then allowed to rest from stimulation for 20 rain before a final moving shadow stimulus was presented, to investigate spontaneous recovery of responses. Fish were tested at the same time of the day on 3 consecutive days.
Histology At the end of the 3 days, those fish which had undergone cranial surgery were killed in a bath of 1 : 5000 MS222 for histological examination of their brains. The cork plugs in the cranium were removed to permit diffusion of 10% formalin fixative into the brain of the decapitated fish. Following 12 h fixation the brains were removed and dehydrated in ethyl and butyl alcohol mixtures followed by butyl alcohol and chloroform treatments prior to wax-embedding. After wax-embedding 15/~m serial sections were cut and mounted prior to staining with hematoxylin-eosin. One sham-operated fish showing dorsal telencephalic damage and two ablated fish, one with only partial posterior telencephalic damage and one with extensive damage to the mesencephalon were excluded from group analysis on the basis of histological examination. Of the remaining 11 telencephalon-ablated fish, all had over 90% of their telencephalon removed on the criteria described in Laming and Ennis (1982). In two of these subjects a small portion of the posterior dorsal telencephalon anterior to the nucleus habenularis remained intact, in the remaining 8 fish the dorsal telencephalon was totally removed to the level of the anterior margin of the nucleus habenularis, situated on the posterior
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
87
telencephalic diencephalic border. Examination of the serial sections from the remaining sham-operated brains revealed no invasive damage or necrosis of the telencephalon or optic tecti. RESULTS Ten N, 7 S, and 10 A fish, with adequate histological confirmation of lesions, were successfully tested on 3 consecutive days and these provided the data for this report. Four other fish were excluded due to movement of the ECG electrodes between Days 2 and 3. All 27 fish showed cardiac and ventilatory deceleration on the initial stimulus presentation on each day though responses to subsequent stimuli on each day habituated (Figs. l a-1 c, 2a-2c). Both initial response magnitude and trials to habituation criterion were analyzed first by a two-way ANOVA with repeated measures on one factor (Days). Differences between groups and days were then further analyzed by the Tukey test for multiple comparisons. Spontaneous recovery and dishabituation to a tap stimulus were assessed by paired t tests.
(i) Initial Responses (a) Cardiac. No significant difference was found in the initial cardiac responses between the three groups [F(2, 24) = 2.94, p = NS] though differences were found between days IF(2, 48) = 7.68, p < .01]. No interaction occurred between groups and days. A decline in initial response occurred over days this being especially evident in the case of group A fish (Table la). There was a tendency, however, for group A fish to have a larger initial cardiac response than that of the controls; this was evident when compared with group N fish on Day 1 (Table la; Figs. la-lc). (b) Ventilatory. Ventilatory responses showed significant differences between groups [F(2, 24) = 3.40, p = < .05] and between days [F(2, 48) = 5.63, p < .01] though there was no interaction effect (Figs. 2a-2c). The decline in response over days was evident when Days 1 and 3 were compared in the N group (Table lb). The higher initial responses of the A group was most evident when it was compared with the N group on Day 3 (Table lb; Fig. 2c).
(2) Response Habituation (a) Cardiac. Significant differences in the habituation of cardiac responses between groups [F(2, 24) = 8.93, p < .01] and between days [F(2, 48) = 24.54, p < .01] occurred though there was no interactive effect. N and A fish showed significantly quicker habituation on Day 3 compared to Day one (Table 2a). No significant differences occurred between control groups on any of the 3 test days, though the A group took significantly longer to habituate on Days 1 and 2 compared with the two control
88
ROONEY AND LAMING
2-42"3l
22-
a
¢ ¢ ; -" "- ,; _- = =
21-
ABLATED SHAM NORMAL
20. g,I
1.9.
P- 1'8. Z
1.7.
=i 1'61,5. Z I-4. O
~ 1"3. re 1-2, 1.1.
--_-_-2
1"0'
. . . . .
-_
. . . . . .
Z-
0"9-
'~ '4 6
8 10 12 le+ 1El 18 20 ~ TRIALS
TO
2 4 ~R
HABITUATION
20]! ° 1,9-~ tu 1.8 i1
¢. : ¢,
ABLATED
"_. -" _"_. " _-
SHAM NORMAL
I-1" m 2: 1.6=E 1-5-
tu
•
1.4-
Z O 1.30,. uJ 1.2iX
1,11,0-
w
_
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
-
0.9-
. 4
. ;
. 8
. 10
TR,ALS
1'2
14
16
18
210
~ -
2
' 4
,~ S.R
TO.ABITUATIO.
FI6. 1. Cardiac responses of goldfish to the repeated presentation of a moving shadow stimulus at 2-min intervals, mean of the ratio of the longest interbeat interval in the stimulus 10 s to that in the prestimulus 10 s. The horizontal solid and dashed lines represent the mean _ SE respectively of this response ratio, in fish prior to the presentation of a stimulus. (a) Test day 1, (b) test day 2, (c) test day 3. T, tap stimulus; SR, spontaneous recovery. ( e ) Ablated, (&) sham, (R) normal.
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
20~
89
c
,81\
-_°=
\
TRIALS
.
TO HABITUATION FI~. 1--Continued.
groups (Table 2a; Fig. la). However, on the subsequent day's testing, between group differences proved nonsignificant though over all days the mean number of trials required for response habituation in the A group was consistently longer than for the controls (Table 2a). (b) Ventilatory. Significant differences in habituation of ventilatory responses between groups [F(2, 24) = 27.51, p < .01] and between days [F(2, 48) = 17.79, p < .01] were apparent. Again no interaction between groups and days was obvious. All fish habituated progressively faster on each day's testing, this being especially evident for the N and A groups (Table 2b). The habituation rates of N and S groups were not significantly different from each other on any of the 3 days. The A group exhibited significantly slower habituation than the S and N groups on all 3 test days (Table 2b; Figs. 2a-2c).
Arousal vs Habituation The magnitude of initial arousal responses to stimuli and the number of those stimuli required for habituation are linearly correlated in normal fish (Laming & Savage, 1980;). In order to determine the effect of ablation on habituation itself, it was therefore necessary to correct for variation in initial response magnitude. This was performed in an empirical manner
90
ROONEY AND LAMING 2.22.1
a e~
-" -" --" ± -" ==:
~O
ABLATED SHAM
1.9UJ 1-8I- - 1"7Z 1'6-
<
1,5MJ 1;4-
Z
1.3~ 1.21-1 1.0 0.9-
S.R TRIALS
_-- ~ _" _- ± _- _- -_
2-0~ 1.9~
TO
HABITUATION
ABLATED SHAM NORMAL
1-8Ci •.n 1-7I•~, 1-81.5IIJ
1.4.
Z 1.30 ~ 1.2wJ 1 l et 1 - 0 - _ _ 0.9-
i
~
~ 1'o 1'2 1', I'B l'S 2'0 ~ TRIALS
TO
~
d
SR
HABITUATION
FIG. 2. Ventilatory responses of goldfish to the repeated presentation of a moving shadow stimulus at 2-min intervals, mean of the ratio of the longest interbreath interval in the stimulus 10 s to that in the prestimulus 10 s. The horizontal solid and dashed lines represent the mean _+ SE respectively of this response ratio in fish prior to the presentation of a stimulus. (a) Test day 1, (b) test day 2, (c) test day 3. T, tap stimulus; SR, spontaneous recovery. ( e ) Ablated, (A) sham, (11) normal.
91
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
20" I,t,I ¢,.. 1.9-
= = = ,I..z x = = --
I-. 1.8Z
ABLATED SHAM NORMAL
1-7-
,,¢ 1-6.
m 1,5, Z
0 1-4. a. tR 1.3uJ ¢
'1.2-
1.1-
i'--
1'C 0.94
~
8
110 112 14
TRIALS
TO
16
1=8 210
TI
~
Z[
I SR
HABITUATION
FIG. 2--Continued.
by calculating the ratio of initial response magnitude to trials to habituation and analyzing the resultant estimates of the slopes of the habituation process (Table 3) (Laming, 1987). (a) Cardiac responses. No difference was found in ECG slope ratio between the groups [F(2, 24) = 1.74, p = NS] though differences were found between days. The group with ablations, however, showed a trend TABLE 1 Initial Response (mean + SE of the mean) of Cardiac (a) and Ventilatory (b) Responses of 10 Normal (N), 7 Sham (S), and 10 Telencephalon-Ablated (A) Goldfish to a Moving Shadow Stimulus on Each of 3 Days Testing Normal (a) Cardiac Day 1 Day 2 Day 3
x 1.67 1.58 1.39
4± ± 4-
SE .13 .16 .27
(b) Ventilatory rate Day 1 Day 2 Day 3
1.90 ± .24 1.46 ± .13 1.24 _+ .07
Tukey
(A)
(A, HI)
Sham x 1.80 1.28 1.32
± ± ± ±
Tukey SE .35 .11 .09
2.08 ± .15 1.72 + .19 1.38 ± .10
Ablated x 2.44 1.65 1.95
_+ ± ± ±
SE .19 .21 .21
2.15 _+ .27 1.82 _+ .21 2.06 -+ .19
Tukey
(N) (H1)
(N)
Note. Tukey, Tukey test for multiple comparisons at the p < .05 level. Group differences with 4, 34 df denoted by the significant comparator group (N, S, or A). Habituation effects with 4, 48 dfdenoted by the significant comparator day (H1, H2 or H3) within the group.
92
ROONEY AND LAMING
TABLE 2 Number of Trials to Habituation (mean ± SE of the mean) of Cardiac (a) and Ventilatory (b) Responses of 10 Normal (N), 7 Sham (S), and 10 Telencephalon-Ablated (A) Goldfish to a Moving Shadow Stimulus on Each Test Day (Significance tests as Table 1) Normal (a) Cardiac Day 1 Day 2 Day 3
x 12.1 8.4 4.8
(b) Ventilatory Day 1 Day 2 Day 3
11 ± .91 8.4 ± .71 4.8 ± .8
Tukey
± SE ± 1.02 (A) ± 1.64 (A) ± .86 (HI)
Sham x 11.71 7.71 7.42
± ± ± ±
Tukey
Ablated + ± ± ±
Tukey
SE 2.04 2.11 1.7
(A) (A)
x 23.3 14.6 10.3
SE 1.93 (N, S) 2.77 (N, S, HI) 1.68 (HI)
(A) 11.57 ± 1.86 (A) 7.57 ± 1.23 (A, HI) 6.0 ± .61
(A) (A) (A)
24.5 ± 1.68 (N, S) 18.7 __- 2.49 (N, S, H1) 15 ± 2.64 (N, S, HI)
toward a more gradual habituation within days compared with the controls. T h e slope o f h a b i t u a t i o n for c a r d i a c r e s p o n s e s i n c r e a s e d significantly o v e r all g r o u p s o v e r d a y s [F(2, 48) = 10.58, p < .01]. W i t h i n g r o u p effects w e r e s e e n for D a y s 1-3 in b o t h N a n d A g r o u p fish (Table 3a). (b) Ventilatory responses. T h e rate of habituation of ventilatory responses m e a s u r e d b y slope was significantly different b e t w e e n both groups [F(2, 24) = 11.41, p < .01] a n d d a y s [F(2, 48) = 7.55, p < .01] t h o u g h n o i n t e r a c t i o n w a s a p p a r e n t . T h e effect o f d a y s was similar to that for c a r d i a c r e s p o n s e d e c l i n e ; viz., the slope i n c r e a s e d with s u c c e s s i v e d a y s , e v i d e n t in N a n d A fish w h e n D a y s 1 a n d 3 w e r e c o m p a r e d . T h e rate o f h a b i t u a t i o n m e a s u r e d b y slope was c o n s i s t e n t l y s t e e p e r for c o n t r o l fish c o m p a r e d to t h o s e with the t e l e n c e p h a l o n a b l a t e d (Table 3b).
TABLE 3 Mean Habituation Slope (mean ± SE of the mean) of (a) Cardiac and (b) Ventilatory Responses of 10 Normal (N), 7 Sham (S), and 10 Telencephalon-Ablated Goldfish to a Moving Shadow Stimulus on Each Test Day (Significance Tests as Table 1) Normal
Tukey
(a) Cardiac Day 1 Day 2 Day 3
x .156 .23 .295
--+_SE ± .01 ± .03 ± .02 (HI)
(b) Ventilatory Day 1 Day 2 Day 3
.170 ± .01 (A) .181 ± .02 .300 ± .03 (A, HI, H2)
Sham x .162 .224 .231
± ± ± ±
Tukey
SE .02 .04 .05
.208 ± .03 .272 ± .07 .239 ± .03
Ablated x .097 .174 .218
(A) (A)
± ± ± ±
SE .01 .05 .03
Tukey
(H1)
.058 ± .003 (S, N) .104 ± .02 (S) .168 ± .03 (N, HI)
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
93
Dishabituation The presentation of a different stimulus, a tap, resulted in recovery of the cardiac response (dishabituation) in all groups for both Days 1 and 2 (Table 4a). Normal and sham-operated fish showed dishabituation of ventilatory responses on both days; however, the group with their telencephalon ablated showed no significant ventilatory response recovery on either day (Table 4b), following tap presentation. On Day 3 the group with their telencephalon ablated showed significant dishabituation of theircardiac and ventilatory responses and the control S and N groups showed a similar trend.
Spontaneous Recovery The N and S groups showed significant spontaneous recovery of ventilatory arousal responses on Days 1 and 2. However, by Day 3 no significant recovery was apparent for either cardiac or ventilatory responses in either group (Table 4). The spontaneous recovery of cardiac responses in the A group was not significant on any of the 3 test days. Significant recovery of ventilatory responses, however, did occur on Day 3.
DISCUSSION Telencephalic ablation did not abolish the ability of fish to habituate cardiac or ventilatory arousal responses. Both control animals and those with their telencephalon ablated showed habituation to repeatedly presented moving shadow stimuli within a test session and all groups showed increased habituation rates on successive days testing. The latter finding supports one of Thompson and Spencer's (1966) parameters for their habituation paradigm which had not previously been examined using the cardiac and ventilatory correlates of behavioural arousal in C. auratus. However, fish with their telencephalon ablated showed significantly slower response habituation within test sessions when compared with their controls, supporting the findings of Laming and Ennis (1982) and Laming and McKee (1981) that telencephalic ablation resulted in disruption of short-term (within session) habituation. The habituation deficit due to ablation was not due solely to the increased responsiveness of the ablated subjects resulting in an increased number of trials to habituation (Laming & Savage, 1980; Rooney & Laming, 1986b), because following a correction for variation in initial response magnitude between groups, the analysis of habituation slopes still showed a consistently more gradual decline in responses (habituation) in telencephalon-ablated subjects. The tendency for increased response magnitude on novel stimulus presentation seen in the present study following telencephalic ablation was not evident in the study of Laming and McKee
94
ROONEY
AND
LAMING
o .,= ,,-;m i Z ii
ca,
r_,:=
ii
~m
,...m I Z ii
tl
Mm
.~ f Z Ii
,-~ i Z
i[
ii
ii
~ I Z
t ~.
ii
Ii v
ii
-4,.-,
,".im
t Z
I Z
t ~,
t ~.
t Z
f o.
ii
ii
II v
ii V
ii
II
II V
it tt
H V
n V
tt H
ii
m~
~
II
~.i ,.~
r_,&
~7
. _,.g_,,,-
H ~I
ii v
+..,
6
6
6
6
['-' rm I q
I q
1 q
u4 q
I q
I Z
II V
II V
II V
II V
II V
II
II
~2
As 5a
I ~.
I N
I Z
I ~.
I ~.
I N
ii v
n
ii
li v
,
~P ii
Jf
it
V
0
~a £a W~ ~b
II V
II V
II V
II V
II
II
8
~a
sa 0
II V
d
E
e-~
eh ~Z ,,,h
~
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
95
(1981). However, the results obtained in their second experiment showed a similar trend. The increased response magnitude on novel stimulus presentation, evident following telencephalic ablation in the present study, suggests the disruption of some inhibitory mechanism on the arousal response. The telencephalon also appears to exert separate inhibitory influences on responses to recently encountered stimuli which results in decreased habituation following telencephalic ablation. This suggests that the telencephalon is important in short-term modulation of arousal. Such an interpretation is similar to the proposed regulatory role for the telencephalon on arousal advanced by Laming and McKee (1981). The present study would also suggest that the predominant telencephalic influence on arousal is inhibitory. Thus, the present findings are in close accord with what would be predicted by the "inhibition of dominant responses" concept which suggests that one function of the telencephalon is in inhibiting dominant behavior patterns or those patterns which have gone unrewarded or punished (responses both learned and innate) (Flood et al., 1976; Flood & Overmier, 1981). Thus, the persistence of cardiac and ventilatory decelerations on stimulus presentation may be considered as a result of arousal, measured by these responses, being dominant. Following ablation of the telencephalon the fish would become less able to inhibit these responses. While indicating habituation deficits within training sessions, the fish with their telencephalon ablated were able to retain some information concerning the previous day's habituation session as all groups showed progressively quicker habituation on each day's testing. This suggests that a lower brain center can register and store sufficient information about the previous training session to facilitate increased habituation. In suggesting no disruption of long-term habituation mechanisms, however, it is noted that fish with their telencephalon ablated appear to show poorer retention of information relating to the eliciting test stimulus from session to session. This was most evident for ventilatory responses where initial response magnitude did not decline significantly on successive days testing. Variation in the results from cardiac recordings which are normally the most consistent correlate of arousal in Carassius (Laming & Savage, 1980) were due to two of the ablated fish showing small initial responses on Days 2 and 3. These fish were retained in group analysis because on the basis of histology both animals had 90% of their telencephalon removed and showed no damage to the other brain structures. Previous studies have implicated the telencephalon with long-term consolidation of memory (Peeke et al., 1972; Gordon, 1979; Peeke & Gordon, 1981; Marino-Neto & Sabbatini, 1983), and while disruption of some long-term memory system is suggested by the present results it appears to be less affected than that involved in short-term (within session) habituation.
96
ROONEY AND LAMING
REFERENCES Flood, N. B., & Overmier, J. B. (1981). Learning in teleost fish: Role of the telencephalon. In P. R. Laming (Ed.), Brain mechanisms of behaviour in lower vertebrates (pp. 259279). Cambridge: Cambridge Univ. Press. Flood, N. B., Overmier, J. B., & Savage, G. E. (1976). Teleost telencephalon and learning: An interpretive review of data and hypotheses. Physiology and Behavior, 16, 783798. Gordon, D. (1979). Effects of forebrain ablation on taste aversion in goldfish (Carassius auratus). Experimental Neurology, 63, 356-366. Laming, P. R. (1980). Electroencephalographic studies on arousal in the goldfish (Carassius auratus). Journal of Comparative and Physiological Psychology, 94, 238-254. Laming, P. R. (1981). The physiological basis of alert behaviour in fish. In P. R. Laming (Ed.), Brain mechanisms of behaviour in lower vertebrates (pp. 203-224). Cambridge, England: Cambridge Univ. Press. Laming, P. R. (1987). Behavioural arousal and its habituation in the squirrel fish Holocentrus rufus; the role of the telencephalon. Behavioural and Neural Biology, 47, 80-104. Laming, P. R., & Ennis, P. (1982). Habituation of fright and arousal responses in the teleosts Carassius auratus and Rutilus rutilus. Journal of Comparative and Physiological Psychology, 96, 460-466. Laming, P. R., & Hornby, P. (1981). The effect of unilateral telencephalic lesions on behavioral arousal and its habituation in the roach, Rutilus rutilus. Behavioral and Neural Biology, 33, 59-65. Laming, P. R., & McKee, M. (1981). Deficits in habituation of cardiac arousal responses incurred by telencephalic ablation in goldfish (Carassius auratus) and their relation to other telencephalic functions. Journal of Comparative and Physiological Psychology, 95, 460-467. Laming, P. R., & Savage, G. E. (1980). Physiological changes observed in the goldfish (Carassius auratus) during behavioural arousal and fright. Behavioral and Neural Biology, 29, 255-57. Laming, P. R., & Savage, G. E. (1981). Seasonal differences in brain activity and responsiveness shown by goldfish (Carassius auratus). Behavioral and Neural Biology, 32, 386-389. Marino-Neto, J., and Sabbatini, R. E. M. (1983). Discrete telencephalic lesions accelerate the habituation rate of behavioral arousal responses in Siamese Fighting fish (Betta splendens). Brazilian Journal of Medical Biological Research, 16, 271-278. Nieuwenhuys, R. (1963). The comparative anatomy of the actinopterygian forebrain. Journal Hirnforsch, 6, 171-192. Peeke, H. V. S., & Gordon, D. G. (1981). Deficits in foraging behavior after telencephalic ablation in the goldfish (Carassius auratus). Experimental Neurology, 72, 204-210. Peeke, V. S., Peeke, S. C., and Williston, J. S. (1972). Long-term memory deficits for habituation of predatory behavior in the forebrain ablated goldfish (Carassius auratus). Experimental Neurology, 38, 288-294. Roberts, M. G., Wright, D. E., and Savage, G. E. (1973). A technique for obtaining the electrocardiogram of fish. Comparative Biochemistry and Physiology A, 44, 665-668. Rooney, D. J., & Laming, P. R. (1986a). Localization of telencephalic regions concerned with habituation of cardiac and ventilatory responses associated with arousal in the goldfish (Carassius auratus). Behavioral Neuroscience, 100(1), 45-50. Rooney, D. J., & Laming, P. R. (1986b). Cardiac and ventilatory arousal responses and their habituation in goldfish. Effects of the intensity of the eliciting stimulus. Physiology and Behavior, 37, 11-t4. Thompson, R. F., & Spencer, W. A. (1966). Habituation: A model phenomenon for the study of neural substrates of behavior. Psychological Review, 73, 16-43.
49, 83-96 (1988)
Effects of Telencephalic Ablation on Habituation of Arousal Responses, within and between Daily Training Sessions in Goldfish D. J. ROONEY AND P. R. LAMING Department of Zoology, The Queen's University of Belfast, Belfast BT7 1NN, United Kingdom Goldfish Carassius auratus were presented with a moving shadow stimulus at 2-min intervals and their cardiac and ventilatory responses were monitored. Normal fish, fish with their telencephalon ablated, and those with sham operations were compared for responsiveness and habituation to repeatedly presented stimuli over the 3-day test period. While all groups showed increased habituation on successive days testing, fish with their telencephalon ablated showed significantly slower response habituation within the daily test sessions when compared with the control groups. Subjects with their telencephalon ablated also showed a tendency for increased responsiveness on initial stimulus presentation and poorer retention between days of information relating to the eliciting test stimulus. © 1988 Academic Press, Inc.
Ablation of the teleostean telencephalon has been shown to impair habituation of nonspecific arousal responses in a number of studies (Laming & McKee, 1981; Laming & Ennis, 1982). Localized lesions in the posterior dorsomedial (Laming & Hornby, 1981) or dorsocentral telencephalic regions (Rooney & Laming, 1986a) have also resulted in significant deficits in habituation of the arousal response of fish to tap or visual stimuli. Extensive damage to the dorsolateral telencephalic areas of the squirrel fish Holocentrus rufus had no effect on habituation of arousal response to visual stimuli though posterior-medial telencephalic damage severely impaired habituation (Laming, 1987). A recent study by Marino-Neto and Sabbatini (1983) reports that lesions to the posterior area dorsalis pars medialis (Dmp) and commissura telencephali dorsalis (CTD) regions of the telencephalon in Siamese fighting fish (Betta splendens) accelerate habituation of arousal responses to tap stimuli: the result opposite to that obtained in Rutilus by Laming & Hornby (1981). Dmp/CTD lesions also resulted in hyperreactivity to novel environmental stimuli in Betta splendens, a finding not demonstrated 83 0163-1047/88 $3.00 Copyright © 1988 by Academic Press, Inc All rights of reproduction in any form reserved.
84
ROONEY AND LAMING
following telencephalic ablation or lesions in Carassius (Laming & McKee, 1981; Laming & Hornby, 1981). The differing findings may be due to species differences, as the CTD is an uncommon structure in actinopterygian fish (Nieuwenhuys, 1963) and its fine connections are unknown. It is also difficult to make close comparisons between Marino-Neto and Sabbatini's (1983) study and earlier work as in their study they investigated combined startle and arousal responses. Habituation of the former is unaffected by telencephalic ablation (Laming & Ennis, 1982). Another reported deficit following ablation of the teleostean telencephalon in species-specific behavioral studies has been in the consolidation of learning. Peeke et al.'s (1972) investigation into the effects of telencephalic ablation on habituation of predatory behavior in Carassius reported no short-term (within session) habituation deficit but fish with their telencephalon ablated showed no savings between days of habituation. Peeke et al. (1972) interpreted these results as evidence of impaired long-term memory but an intact short-term memory following telencephalic ablation. Similar disruption of long-term memory, resulting in deficits in foraging behavior after telencephalic ablation, in Carassius have also been found by Peeke and Gordon (1981). Gordon (1979) also reports disruption of memory following telencephalic ablation in a taste aversion study. Goldfish with their telencephalon ablated were unable to retain a taste-illness association from one trial to the next to avoid illness associated with LiC1 injection. Increased habituation on repeated testing is one of the parametric characteristics of habituation (Thompson & Spencer, 1966). Thus, an investigation into the effects of telencephalic ablation on habituation of the physiological correlates of the behavioral arousal response during successive days testing should clarify telencephalic influences on shortterm and long-term habituation. It should also address both the arousal and memory hypotheses for telencephalic function. This was the rationale for the present experiment in which the habituation of cardiac and ventilatory arousal responses to a moving shadow stimulus in Carassius were studied. MATERIALS AND METHODS Subjects
Goldfish, Carassius auratus, were obtained from a local supplier. They were maintained in 2 × 0.5 × 0.5-m glass aquaria of aerated filtered water at 12 _+ 2°C for 4 weeks prior to experiments. These took place between April and early June to avoid any seasonal variation in responsiveness (Laming & Savage, 1981). Thirty-four animals, 9-12 cm in length were randomly selected from stock tanks and assigned to one of three groups with differing degrees of surgical treatment. These were (1)
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
85
fish with the telencephalic lobes exposed and forebrain ablated by aspiration (A); (2) fish with the forebrain exposed and the telencephalic lobes left intact, the sham operated (S) group; and (3) fish with no cranial surgery, the normal (N) group.
ECG and Ventilatory Recording Procedures The ECG and ventilatory recording procedures were as described in Laming and Savage (1980). Briefly, fish were anesthetized by immersion in a solution of 1 : 10,000 MS222 (tricaine methane sulphonate) until opercular movements ceased. All fish were implanted with electrocardiogram (ECG) electrodes using the method of Roberts et al. (1973). A ventilatory catheter connecting the buccal cavity to a pressure transducer was also implanted. ECGs and ventilations were monitored by a Beckman R411 dynograph.
Cranial Surgery (I) Ablations (A). The 13 fish randomly selected for forebrain ablations were placed in the fish clamp, and their gills were perfused with a 1 : 20,000 solution of MS222 via a tube to the mouth. A 3-mm diameter observation hole was opened in the cranium midway between the eyes. Aspiration of the fatty cerebrospinal fluid exposed the telencephalic lobes which were then ablated by aspiration. Following telencephalic ablation the cranial cavity was filled with teleost saline and the inspection hole was sealed with a sterile cork and Eastman 910 adhesive (Laming, 1980). (2) Sham operation (S). The 10 sham operated fish underwent identical surgery though the telencephalic lobes were left intact. (3) " N o r m a l " fish (N) had no cranial surgery and acted as controls for the sham operation. All fish were revived by perfusion of their gills with fresh water, and allowed to recover in stock aquaria for 48 h prior to testing.
The Experimental Situation and Stimuli After recovery, fish were placed in a 22 x 6 x 10-cm-high wire trough placed centrally in a 35 x 25 x 25-cm-high perspex aquarium containing 12 cm of aerated water at 12 _+ 2°C. The aquarium was mounted on sponge rubber and situated in a 46 x 36 x 26-cm-high black box. The lid of the box contained the apparatus for providing the moving shadow and a longitudinally placed 60 W tungsten filament strip light gave a background illumination of 90 lux. A black screen (35 x 25 cm) fitted below the light incorporated a 10 x 5-cm central aperture. A rotating 22-cm diameter clear Perspex disk, split into six segments, alternatively clear and dark, passed between the light source and screen aperture at a speed of 0.7 cm/s. The stimulus was automatically presented for 10 s in every 120 s, allowing one dark segment to pass the aperture during
86
ROONEY AND LAMING
a stimulus presentation, starting and ending in a clear field. The precise position at which the stimulus ended was controlled by a photocell. The moving shadow approached the fish's visual field from behind as this has been shown to consistently elicit arousal responses (Laming & Savage, 1980). The fish were left for 1 h in the test tank to adapt to the situation prior to the presentation of stimuli. At the end of this period, spontaneous cardiac and ventilatory rate changes were monitored during two consecutive 10-s periods, to provide a baseline for responses to the stimuli which were then serially presented. Habituation of a response in an individual fish was considered to have occurred on the second of two consecutive stimulus presentations in which the ratio of the longest interbeat/interbreath interval in the 10-s stimulus period compared with that in the 10-s prestimulus period fell within the 95% confidence limits of the control (no stimulus) variation from unity (Laming & McKee, 1981). Following at least 20 stimulus presentations, fish were given a tap stimulus for one trial to investigate response dishabituation. The tap was delivered to the side of the box by a Perspex rod mounted on a remotely operated 24 V dc solenoid. Following presentation of the tap, resumption of stimulation with the moving shadow continued for five more trials. The fish were then allowed to rest from stimulation for 20 rain before a final moving shadow stimulus was presented, to investigate spontaneous recovery of responses. Fish were tested at the same time of the day on 3 consecutive days.
Histology At the end of the 3 days, those fish which had undergone cranial surgery were killed in a bath of 1 : 5000 MS222 for histological examination of their brains. The cork plugs in the cranium were removed to permit diffusion of 10% formalin fixative into the brain of the decapitated fish. Following 12 h fixation the brains were removed and dehydrated in ethyl and butyl alcohol mixtures followed by butyl alcohol and chloroform treatments prior to wax-embedding. After wax-embedding 15/~m serial sections were cut and mounted prior to staining with hematoxylin-eosin. One sham-operated fish showing dorsal telencephalic damage and two ablated fish, one with only partial posterior telencephalic damage and one with extensive damage to the mesencephalon were excluded from group analysis on the basis of histological examination. Of the remaining 11 telencephalon-ablated fish, all had over 90% of their telencephalon removed on the criteria described in Laming and Ennis (1982). In two of these subjects a small portion of the posterior dorsal telencephalon anterior to the nucleus habenularis remained intact, in the remaining 8 fish the dorsal telencephalon was totally removed to the level of the anterior margin of the nucleus habenularis, situated on the posterior
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
87
telencephalic diencephalic border. Examination of the serial sections from the remaining sham-operated brains revealed no invasive damage or necrosis of the telencephalon or optic tecti. RESULTS Ten N, 7 S, and 10 A fish, with adequate histological confirmation of lesions, were successfully tested on 3 consecutive days and these provided the data for this report. Four other fish were excluded due to movement of the ECG electrodes between Days 2 and 3. All 27 fish showed cardiac and ventilatory deceleration on the initial stimulus presentation on each day though responses to subsequent stimuli on each day habituated (Figs. l a-1 c, 2a-2c). Both initial response magnitude and trials to habituation criterion were analyzed first by a two-way ANOVA with repeated measures on one factor (Days). Differences between groups and days were then further analyzed by the Tukey test for multiple comparisons. Spontaneous recovery and dishabituation to a tap stimulus were assessed by paired t tests.
(i) Initial Responses (a) Cardiac. No significant difference was found in the initial cardiac responses between the three groups [F(2, 24) = 2.94, p = NS] though differences were found between days IF(2, 48) = 7.68, p < .01]. No interaction occurred between groups and days. A decline in initial response occurred over days this being especially evident in the case of group A fish (Table la). There was a tendency, however, for group A fish to have a larger initial cardiac response than that of the controls; this was evident when compared with group N fish on Day 1 (Table la; Figs. la-lc). (b) Ventilatory. Ventilatory responses showed significant differences between groups [F(2, 24) = 3.40, p = < .05] and between days [F(2, 48) = 5.63, p < .01] though there was no interaction effect (Figs. 2a-2c). The decline in response over days was evident when Days 1 and 3 were compared in the N group (Table lb). The higher initial responses of the A group was most evident when it was compared with the N group on Day 3 (Table lb; Fig. 2c).
(2) Response Habituation (a) Cardiac. Significant differences in the habituation of cardiac responses between groups [F(2, 24) = 8.93, p < .01] and between days [F(2, 48) = 24.54, p < .01] occurred though there was no interactive effect. N and A fish showed significantly quicker habituation on Day 3 compared to Day one (Table 2a). No significant differences occurred between control groups on any of the 3 test days, though the A group took significantly longer to habituate on Days 1 and 2 compared with the two control
88
ROONEY AND LAMING
2-42"3l
22-
a
¢ ¢ ; -" "- ,; _- = =
21-
ABLATED SHAM NORMAL
20. g,I
1.9.
P- 1'8. Z
1.7.
=i 1'61,5. Z I-4. O
~ 1"3. re 1-2, 1.1.
--_-_-2
1"0'
. . . . .
-_
. . . . . .
Z-
0"9-
'~ '4 6
8 10 12 le+ 1El 18 20 ~ TRIALS
TO
2 4 ~R
HABITUATION
20]! ° 1,9-~ tu 1.8 i1
¢. : ¢,
ABLATED
"_. -" _"_. " _-
SHAM NORMAL
I-1" m 2: 1.6=E 1-5-
tu
•
1.4-
Z O 1.30,. uJ 1.2iX
1,11,0-
w
_
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
-
0.9-
. 4
. ;
. 8
. 10
TR,ALS
1'2
14
16
18
210
~ -
2
' 4
,~ S.R
TO.ABITUATIO.
FI6. 1. Cardiac responses of goldfish to the repeated presentation of a moving shadow stimulus at 2-min intervals, mean of the ratio of the longest interbeat interval in the stimulus 10 s to that in the prestimulus 10 s. The horizontal solid and dashed lines represent the mean _ SE respectively of this response ratio, in fish prior to the presentation of a stimulus. (a) Test day 1, (b) test day 2, (c) test day 3. T, tap stimulus; SR, spontaneous recovery. ( e ) Ablated, (&) sham, (R) normal.
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
20~
89
c
,81\
-_°=
\
TRIALS
.
TO HABITUATION FI~. 1--Continued.
groups (Table 2a; Fig. la). However, on the subsequent day's testing, between group differences proved nonsignificant though over all days the mean number of trials required for response habituation in the A group was consistently longer than for the controls (Table 2a). (b) Ventilatory. Significant differences in habituation of ventilatory responses between groups [F(2, 24) = 27.51, p < .01] and between days [F(2, 48) = 17.79, p < .01] were apparent. Again no interaction between groups and days was obvious. All fish habituated progressively faster on each day's testing, this being especially evident for the N and A groups (Table 2b). The habituation rates of N and S groups were not significantly different from each other on any of the 3 days. The A group exhibited significantly slower habituation than the S and N groups on all 3 test days (Table 2b; Figs. 2a-2c).
Arousal vs Habituation The magnitude of initial arousal responses to stimuli and the number of those stimuli required for habituation are linearly correlated in normal fish (Laming & Savage, 1980;). In order to determine the effect of ablation on habituation itself, it was therefore necessary to correct for variation in initial response magnitude. This was performed in an empirical manner
90
ROONEY AND LAMING 2.22.1
a e~
-" -" --" ± -" ==:
~O
ABLATED SHAM
1.9UJ 1-8I- - 1"7Z 1'6-
<
1,5MJ 1;4-
Z
1.3~ 1.21-1 1.0 0.9-
S.R TRIALS
_-- ~ _" _- ± _- _- -_
2-0~ 1.9~
TO
HABITUATION
ABLATED SHAM NORMAL
1-8Ci •.n 1-7I•~, 1-81.5IIJ
1.4.
Z 1.30 ~ 1.2wJ 1 l et 1 - 0 - _ _ 0.9-
i
~
~ 1'o 1'2 1', I'B l'S 2'0 ~ TRIALS
TO
~
d
SR
HABITUATION
FIG. 2. Ventilatory responses of goldfish to the repeated presentation of a moving shadow stimulus at 2-min intervals, mean of the ratio of the longest interbreath interval in the stimulus 10 s to that in the prestimulus 10 s. The horizontal solid and dashed lines represent the mean _+ SE respectively of this response ratio in fish prior to the presentation of a stimulus. (a) Test day 1, (b) test day 2, (c) test day 3. T, tap stimulus; SR, spontaneous recovery. ( e ) Ablated, (A) sham, (11) normal.
91
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
20" I,t,I ¢,.. 1.9-
= = = ,I..z x = = --
I-. 1.8Z
ABLATED SHAM NORMAL
1-7-
,,¢ 1-6.
m 1,5, Z
0 1-4. a. tR 1.3uJ ¢
'1.2-
1.1-
i'--
1'C 0.94
~
8
110 112 14
TRIALS
TO
16
1=8 210
TI
~
Z[
I SR
HABITUATION
FIG. 2--Continued.
by calculating the ratio of initial response magnitude to trials to habituation and analyzing the resultant estimates of the slopes of the habituation process (Table 3) (Laming, 1987). (a) Cardiac responses. No difference was found in ECG slope ratio between the groups [F(2, 24) = 1.74, p = NS] though differences were found between days. The group with ablations, however, showed a trend TABLE 1 Initial Response (mean + SE of the mean) of Cardiac (a) and Ventilatory (b) Responses of 10 Normal (N), 7 Sham (S), and 10 Telencephalon-Ablated (A) Goldfish to a Moving Shadow Stimulus on Each of 3 Days Testing Normal (a) Cardiac Day 1 Day 2 Day 3
x 1.67 1.58 1.39
4± ± 4-
SE .13 .16 .27
(b) Ventilatory rate Day 1 Day 2 Day 3
1.90 ± .24 1.46 ± .13 1.24 _+ .07
Tukey
(A)
(A, HI)
Sham x 1.80 1.28 1.32
± ± ± ±
Tukey SE .35 .11 .09
2.08 ± .15 1.72 + .19 1.38 ± .10
Ablated x 2.44 1.65 1.95
_+ ± ± ±
SE .19 .21 .21
2.15 _+ .27 1.82 _+ .21 2.06 -+ .19
Tukey
(N) (H1)
(N)
Note. Tukey, Tukey test for multiple comparisons at the p < .05 level. Group differences with 4, 34 df denoted by the significant comparator group (N, S, or A). Habituation effects with 4, 48 dfdenoted by the significant comparator day (H1, H2 or H3) within the group.
92
ROONEY AND LAMING
TABLE 2 Number of Trials to Habituation (mean ± SE of the mean) of Cardiac (a) and Ventilatory (b) Responses of 10 Normal (N), 7 Sham (S), and 10 Telencephalon-Ablated (A) Goldfish to a Moving Shadow Stimulus on Each Test Day (Significance tests as Table 1) Normal (a) Cardiac Day 1 Day 2 Day 3
x 12.1 8.4 4.8
(b) Ventilatory Day 1 Day 2 Day 3
11 ± .91 8.4 ± .71 4.8 ± .8
Tukey
± SE ± 1.02 (A) ± 1.64 (A) ± .86 (HI)
Sham x 11.71 7.71 7.42
± ± ± ±
Tukey
Ablated + ± ± ±
Tukey
SE 2.04 2.11 1.7
(A) (A)
x 23.3 14.6 10.3
SE 1.93 (N, S) 2.77 (N, S, HI) 1.68 (HI)
(A) 11.57 ± 1.86 (A) 7.57 ± 1.23 (A, HI) 6.0 ± .61
(A) (A) (A)
24.5 ± 1.68 (N, S) 18.7 __- 2.49 (N, S, H1) 15 ± 2.64 (N, S, HI)
toward a more gradual habituation within days compared with the controls. T h e slope o f h a b i t u a t i o n for c a r d i a c r e s p o n s e s i n c r e a s e d significantly o v e r all g r o u p s o v e r d a y s [F(2, 48) = 10.58, p < .01]. W i t h i n g r o u p effects w e r e s e e n for D a y s 1-3 in b o t h N a n d A g r o u p fish (Table 3a). (b) Ventilatory responses. T h e rate of habituation of ventilatory responses m e a s u r e d b y slope was significantly different b e t w e e n both groups [F(2, 24) = 11.41, p < .01] a n d d a y s [F(2, 48) = 7.55, p < .01] t h o u g h n o i n t e r a c t i o n w a s a p p a r e n t . T h e effect o f d a y s was similar to that for c a r d i a c r e s p o n s e d e c l i n e ; viz., the slope i n c r e a s e d with s u c c e s s i v e d a y s , e v i d e n t in N a n d A fish w h e n D a y s 1 a n d 3 w e r e c o m p a r e d . T h e rate o f h a b i t u a t i o n m e a s u r e d b y slope was c o n s i s t e n t l y s t e e p e r for c o n t r o l fish c o m p a r e d to t h o s e with the t e l e n c e p h a l o n a b l a t e d (Table 3b).
TABLE 3 Mean Habituation Slope (mean ± SE of the mean) of (a) Cardiac and (b) Ventilatory Responses of 10 Normal (N), 7 Sham (S), and 10 Telencephalon-Ablated Goldfish to a Moving Shadow Stimulus on Each Test Day (Significance Tests as Table 1) Normal
Tukey
(a) Cardiac Day 1 Day 2 Day 3
x .156 .23 .295
--+_SE ± .01 ± .03 ± .02 (HI)
(b) Ventilatory Day 1 Day 2 Day 3
.170 ± .01 (A) .181 ± .02 .300 ± .03 (A, HI, H2)
Sham x .162 .224 .231
± ± ± ±
Tukey
SE .02 .04 .05
.208 ± .03 .272 ± .07 .239 ± .03
Ablated x .097 .174 .218
(A) (A)
± ± ± ±
SE .01 .05 .03
Tukey
(H1)
.058 ± .003 (S, N) .104 ± .02 (S) .168 ± .03 (N, HI)
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
93
Dishabituation The presentation of a different stimulus, a tap, resulted in recovery of the cardiac response (dishabituation) in all groups for both Days 1 and 2 (Table 4a). Normal and sham-operated fish showed dishabituation of ventilatory responses on both days; however, the group with their telencephalon ablated showed no significant ventilatory response recovery on either day (Table 4b), following tap presentation. On Day 3 the group with their telencephalon ablated showed significant dishabituation of theircardiac and ventilatory responses and the control S and N groups showed a similar trend.
Spontaneous Recovery The N and S groups showed significant spontaneous recovery of ventilatory arousal responses on Days 1 and 2. However, by Day 3 no significant recovery was apparent for either cardiac or ventilatory responses in either group (Table 4). The spontaneous recovery of cardiac responses in the A group was not significant on any of the 3 test days. Significant recovery of ventilatory responses, however, did occur on Day 3.
DISCUSSION Telencephalic ablation did not abolish the ability of fish to habituate cardiac or ventilatory arousal responses. Both control animals and those with their telencephalon ablated showed habituation to repeatedly presented moving shadow stimuli within a test session and all groups showed increased habituation rates on successive days testing. The latter finding supports one of Thompson and Spencer's (1966) parameters for their habituation paradigm which had not previously been examined using the cardiac and ventilatory correlates of behavioural arousal in C. auratus. However, fish with their telencephalon ablated showed significantly slower response habituation within test sessions when compared with their controls, supporting the findings of Laming and Ennis (1982) and Laming and McKee (1981) that telencephalic ablation resulted in disruption of short-term (within session) habituation. The habituation deficit due to ablation was not due solely to the increased responsiveness of the ablated subjects resulting in an increased number of trials to habituation (Laming & Savage, 1980; Rooney & Laming, 1986b), because following a correction for variation in initial response magnitude between groups, the analysis of habituation slopes still showed a consistently more gradual decline in responses (habituation) in telencephalon-ablated subjects. The tendency for increased response magnitude on novel stimulus presentation seen in the present study following telencephalic ablation was not evident in the study of Laming and McKee
94
ROONEY
AND
LAMING
o .,= ,,-;m i Z ii
ca,
r_,:=
ii
~m
,...m I Z ii
tl
Mm
.~ f Z Ii
,-~ i Z
i[
ii
ii
~ I Z
t ~.
ii
Ii v
ii
-4,.-,
,".im
t Z
I Z
t ~,
t ~.
t Z
f o.
ii
ii
II v
ii V
ii
II
II V
it tt
H V
n V
tt H
ii
m~
~
II
~.i ,.~
r_,&
~7
. _,.g_,,,-
H ~I
ii v
+..,
6
6
6
6
['-' rm I q
I q
1 q
u4 q
I q
I Z
II V
II V
II V
II V
II V
II
II
~2
As 5a
I ~.
I N
I Z
I ~.
I ~.
I N
ii v
n
ii
li v
,
~P ii
Jf
it
V
0
~a £a W~ ~b
II V
II V
II V
II V
II
II
8
~a
sa 0
II V
d
E
e-~
eh ~Z ,,,h
~
EFFECTS OF TELENCEPHALIC ABLATION IN GOLDFISH
95
(1981). However, the results obtained in their second experiment showed a similar trend. The increased response magnitude on novel stimulus presentation, evident following telencephalic ablation in the present study, suggests the disruption of some inhibitory mechanism on the arousal response. The telencephalon also appears to exert separate inhibitory influences on responses to recently encountered stimuli which results in decreased habituation following telencephalic ablation. This suggests that the telencephalon is important in short-term modulation of arousal. Such an interpretation is similar to the proposed regulatory role for the telencephalon on arousal advanced by Laming and McKee (1981). The present study would also suggest that the predominant telencephalic influence on arousal is inhibitory. Thus, the present findings are in close accord with what would be predicted by the "inhibition of dominant responses" concept which suggests that one function of the telencephalon is in inhibiting dominant behavior patterns or those patterns which have gone unrewarded or punished (responses both learned and innate) (Flood et al., 1976; Flood & Overmier, 1981). Thus, the persistence of cardiac and ventilatory decelerations on stimulus presentation may be considered as a result of arousal, measured by these responses, being dominant. Following ablation of the telencephalon the fish would become less able to inhibit these responses. While indicating habituation deficits within training sessions, the fish with their telencephalon ablated were able to retain some information concerning the previous day's habituation session as all groups showed progressively quicker habituation on each day's testing. This suggests that a lower brain center can register and store sufficient information about the previous training session to facilitate increased habituation. In suggesting no disruption of long-term habituation mechanisms, however, it is noted that fish with their telencephalon ablated appear to show poorer retention of information relating to the eliciting test stimulus from session to session. This was most evident for ventilatory responses where initial response magnitude did not decline significantly on successive days testing. Variation in the results from cardiac recordings which are normally the most consistent correlate of arousal in Carassius (Laming & Savage, 1980) were due to two of the ablated fish showing small initial responses on Days 2 and 3. These fish were retained in group analysis because on the basis of histology both animals had 90% of their telencephalon removed and showed no damage to the other brain structures. Previous studies have implicated the telencephalon with long-term consolidation of memory (Peeke et al., 1972; Gordon, 1979; Peeke & Gordon, 1981; Marino-Neto & Sabbatini, 1983), and while disruption of some long-term memory system is suggested by the present results it appears to be less affected than that involved in short-term (within session) habituation.
96
ROONEY AND LAMING
REFERENCES Flood, N. B., & Overmier, J. B. (1981). Learning in teleost fish: Role of the telencephalon. In P. R. Laming (Ed.), Brain mechanisms of behaviour in lower vertebrates (pp. 259279). Cambridge: Cambridge Univ. Press. Flood, N. B., Overmier, J. B., & Savage, G. E. (1976). Teleost telencephalon and learning: An interpretive review of data and hypotheses. Physiology and Behavior, 16, 783798. Gordon, D. (1979). Effects of forebrain ablation on taste aversion in goldfish (Carassius auratus). Experimental Neurology, 63, 356-366. Laming, P. R. (1980). Electroencephalographic studies on arousal in the goldfish (Carassius auratus). Journal of Comparative and Physiological Psychology, 94, 238-254. Laming, P. R. (1981). The physiological basis of alert behaviour in fish. In P. R. Laming (Ed.), Brain mechanisms of behaviour in lower vertebrates (pp. 203-224). Cambridge, England: Cambridge Univ. Press. Laming, P. R. (1987). Behavioural arousal and its habituation in the squirrel fish Holocentrus rufus; the role of the telencephalon. Behavioural and Neural Biology, 47, 80-104. Laming, P. R., & Ennis, P. (1982). Habituation of fright and arousal responses in the teleosts Carassius auratus and Rutilus rutilus. Journal of Comparative and Physiological Psychology, 96, 460-466. Laming, P. R., & Hornby, P. (1981). The effect of unilateral telencephalic lesions on behavioral arousal and its habituation in the roach, Rutilus rutilus. Behavioral and Neural Biology, 33, 59-65. Laming, P. R., & McKee, M. (1981). Deficits in habituation of cardiac arousal responses incurred by telencephalic ablation in goldfish (Carassius auratus) and their relation to other telencephalic functions. Journal of Comparative and Physiological Psychology, 95, 460-467. Laming, P. R., & Savage, G. E. (1980). Physiological changes observed in the goldfish (Carassius auratus) during behavioural arousal and fright. Behavioral and Neural Biology, 29, 255-57. Laming, P. R., & Savage, G. E. (1981). Seasonal differences in brain activity and responsiveness shown by goldfish (Carassius auratus). Behavioral and Neural Biology, 32, 386-389. Marino-Neto, J., and Sabbatini, R. E. M. (1983). Discrete telencephalic lesions accelerate the habituation rate of behavioral arousal responses in Siamese Fighting fish (Betta splendens). Brazilian Journal of Medical Biological Research, 16, 271-278. Nieuwenhuys, R. (1963). The comparative anatomy of the actinopterygian forebrain. Journal Hirnforsch, 6, 171-192. Peeke, H. V. S., & Gordon, D. G. (1981). Deficits in foraging behavior after telencephalic ablation in the goldfish (Carassius auratus). Experimental Neurology, 72, 204-210. Peeke, V. S., Peeke, S. C., and Williston, J. S. (1972). Long-term memory deficits for habituation of predatory behavior in the forebrain ablated goldfish (Carassius auratus). Experimental Neurology, 38, 288-294. Roberts, M. G., Wright, D. E., and Savage, G. E. (1973). A technique for obtaining the electrocardiogram of fish. Comparative Biochemistry and Physiology A, 44, 665-668. Rooney, D. J., & Laming, P. R. (1986a). Localization of telencephalic regions concerned with habituation of cardiac and ventilatory responses associated with arousal in the goldfish (Carassius auratus). Behavioral Neuroscience, 100(1), 45-50. Rooney, D. J., & Laming, P. R. (1986b). Cardiac and ventilatory arousal responses and their habituation in goldfish. Effects of the intensity of the eliciting stimulus. Physiology and Behavior, 37, 11-t4. Thompson, R. F., & Spencer, W. A. (1966). Habituation: A model phenomenon for the study of neural substrates of behavior. Psychological Review, 73, 16-43.