Ontogeny of Active Avoidance in the Rat: Learning and Memory JAROMIR M Y S L I V E ~ E K Institute of Hygiene and Epidemiology Praha 10, Czechoslovakia JARMILA HASSMANNOVA Department of Physiology Medical Faculty of Hygiene Charles University Praha, Czechoslovukiu Ontogenetic development of active avoidance learning, extinction and retention was studied in rats. The learning of a 1-way active avoidance was most rapid between Weeks 4 and 6 , although some slight gender-related differences were evident. No such unambiguous development was detected in forced extinction. The 24-hr retention of avoidance peaked at the age of 4 weeks whereas 1-month retention was best in animals trained at the age of 8 weeks. The retrieval of memory trace also had best values at these ages. Retention of forced extinction was found to peak in 6-week animals. The existence of developmental “critical periods” must be considered cautiously as various functions have different time courses depending upon the chosen parameters in assessment.
Various aspects of the ontogenetic development of learning and memory have been studied on different occasions. Primitive forms of learning appear very early in the ontogeny: in altricial animals they appear practically immediately after birth; in precocial animals simple forms of conditioned reflexes appear before delivery or hatching (Caldwell & Werboff, 1962; Gray, Yates, & McNeal, 1967; Sedlicek, 1963; Thoman, Wetzel, & Levine, 1968; Troshikhin, 1952; for review see Volokhov, 1970). The possibility of conditioning depends on the mutual relations between biological importance (motivational systems) of conditioned and unconditioned stimuli, on their time sequence, and on the maturational characteristics of the functional systems participating in conditioning, as has been already shown (Campbell & Spear, 1972; Myslivecek, 1972). Active 1-way avoidance may be considered as an artificial situation modeling important events of the daily life: the signals and cues guiding this defense reaction need to be learned and remembered: in cats and dogs, for example, escape and active avoidance learning to somesthetic-kinesthetic stimuli has been established before eye-opening (9th day; Bacon, 1973; Bacon & Stanley, 1968; Stanley, Barret, & Bacon,
Reprint requests should be sent to Professor Jaromir Myslivecck, M.D., D.Sc., IHE, Srobarova 48, 10042 Praha 10, Czechoslovakia. Received for publication 14 April 1977 Revised for publication 1 December 1977 Developmental Psychobiology, 12(2): 169-186 (1 979) @ 1979 by John Wiley & Sons, Inc.
0012-1630/79/0012-0169$01.OO
170
MYSLIVECEK AND HASSMANNOVA
1974). Although several papers have dealt with the ontogenetic development of active avoidance in rats covering different periods of life from 19 days (Riccio, Rohrbaugh, & Hodges, 1968) up to 2 years (Dobrovolskaya, 1972; Doty, 1966a,b), no general agreement has been reached as to the ontogeny of active avoidance learning. Some authors deny any changes in learning during ontogeny (Feigley & Spear, 1970; Kirby, 1963; Porter & Thompson, 1967; Riccio et al., 1968). On the other hand, Deneberg and Kline (1958) have described better learning in 60-day rats as compared with 30and 225-day-old animals. Egger and Livesey (1972) have found a monotonically decreasing number of trials to criterion with age in animals at the age of 25, 50, and 100 days. Obraztsova, Dobrovolskaya, Fedorov, and Bogdanova (1972) have: stressed an “exaltation period” of learning and other neural functions between the 1st and 6th month of the rat’s life. Izquierdo, Salzano, ThaddCu, and ThomC (1975) have given an account of better learning in 55-70-day-old rats than in weanlings (20-21 days old) as, similarly, have McLaughlin, Eller, and Korol (1975) in 120-day-old rats compared with 21- and 480-day-old animals. An impairment in old animals is reported also by others (Doty, 1966a,b; Fields, 1953; Freund and Walker, 1971). On the other hand, common agreement prevails in the literature that memory (not only in active avoidance) operates at a higher level in adults than in infant or weanling animals (see Campbell & Spear, 1972; Myslivecek, 1976). Most of the mentioned studies are unfortunately rather episodic, as only some representatives of different developmental periods were investigated. To overcome this gap we have studied the development of avoidance learning and remembering more systematically, starting with the age of 2 weeks up to 3 months (Myslivecek & Hassmannovi, 1973). We have found that optimum 1-way avoidance learning lies between Weeks 4 and 6. The present paper extends and complements our previous work. It deals with different criteria of learning and memory. The aim of the experimentation was to obtain data on active avoidance learning and remembering at precisely delimited ages and to cover with this investigation the lifespan from suckling period (2 weeks) up to adult age ( 3 months). The changes of memory capacity were investigated in 1-day and 1-month retention. The rat pup is an altricial animal with a considerable after-delivery delay before reaching some important degree of early neural and sensory maturation, which happens at the age of about 2 weeks when, among other things, the eyelids open and the acoustic meatuses become unplugged. To avoid developmental shifts at every selected age period the animals had to meet the learning criterion within 1 or 2 experimental days. Thus, we compared animals of increasing age with 1 week intervals up to 6 weeks and then at 8 weeks and at 3 months. A further aim of this study was to find out potential gender-related changes in learning development. This question has an old history (see, for a review of early work, the paper by Tomilin & Stone, 1933). More recently gender differences in avoidance learning (Beatty & Beatty, 1970; Denti & Epstein, 1972; Dobrovolskaya, 1973), development of pyramidal cells in the sensorimotor cortex (Gregory, 1975), maturation of open field behavior (Valley & Boes, 1976), sensitivity to electric shock (Beatty & Fessler, 1976), and ontogeny of self-stimulation behavior (Welley, Stinus, & Cardo, 1975) have been described.
ONTOGENY OF ACTIVE AVOIDANCE
17 1
Methods Materials Albino rats (Rattus norvegicus, Wistar strain) of both genders aged 2,3,4,5,6, and 8 weeks and 3 months, bred in the animal house of the Institute of Physiology, Medical Faculty, Charles University in Prague, were used in the experiment. A total of 146 animals was studied. The animals in each age and gender group (minimum, 10 animals per group) were offspring randomly taken from 4-6 litters.
Apparatus The animals were trained in an automatic 1-way avoidance apparatus described earlier (Myslivecek, Zalud, Dallinger, & Hassmannovi, 1970). Briefly, the apparatus consisted of 2 platforms (one of them electrified) separated during the intertrial intervals (ITI) by a sliding panel. In the present experiment both platforms were flush to equalize the difficulty of experimental task for different ages as far as possible. The strength of the current applied to the electrified floor serving as unconditioned stimulus (US) was adjusted individually for each animal to overreach, just slightly, the threshold levels. The strength of unscrambled footshock which was estimated preliminarily for each age group was changed stepwise within the 1st 2-3 trials, if necessary. Simultaneously with the conditioned stimulus (CS)-a white light on (located on the panel)-the panel was moved to the end of the apparatus and uncovered the safe platform (thus, the CS had a complex character: light and movement of the panel). The C S lasted 15 sec: 5 sec of isolate application, reinforced for another 10 sec by the US. The IT1 duration was 25 sec. The safe platform was accessible during the whole CS and US action and during the 1st 5 sec of IT1 (Fig. 1A). When the program of extinction was begun, the animal was automatically removed by the movement of the panel as soon as it stepped on the safe platform; therefore, we used the term “forced” extinction. The panel then again returned and further access to the safe platform was made possible.
Procedure The experiment started when the animal reached the age of 14,21,. . . days of life, precisely. Of the following 6 experimental tasks all but the last one were studied within 3-4 days (Fig. IB): 1. 1st day-establishment of the avoidance reaction to 1 or both of the 2 criteria to be met (a basic one, 9 correct responses-avoidances-in 10 subsequent trials, and a more severe criterion, 9 consecutive correct responses; only the basic criterion of this 1st task was attempted again the next day if the training on the 1st day was not successful; all other criteria were attempted only once); 2. 2nd (or 3rd) day-a 24-hr retention of the avoidance reaction (10 trials with reinforcement); 3. after 5-min break-forced extinction of avoidance reaction to the same criteria
172
MYSLIVECEK AND HASSMANNOVA
-
A I
cs
:
'
IS'
us
'
18'
:
us
'
MWO 1
cs
' 4
20'
r
:
1
MWO
IS'
'
I
I T 1
I T 1
c----+---t--e---+-
4
B day:
I.
II.
111.
xxx.
Avoidance trammg to criterion
: 4
I
0,
upto 30 trials
Fig. 1. (A) Schedule of CS and US presentation; accessibility of the safe platform initiated and terminated by movements of the sliding wall (MWO) and ITI. (B) Schedule of the experiment.
as with learning (i.e. 90% correct responses within 10 trials and 9 conseculive correct responses); 4. 3rd (or 4th) day-a 24-hr retention of the extinguished avoidance (10 trials without reinforcement); 5. after 5-min break-relearning of avoidance now to a criterion of 75% of correct responses in 20 consecutive trials; 6 . after 4 weeks-a 1-month retention (20 trials with reinforcement). The fulfilment of each criterion had to begin before the 30th trial. If the animal failed to continue with correct performance no further attempts were made to attain the criteria of corresponding task. Only failure at the easier criterion (9/10) stopped further work with the animal. Meeting the more severe criterion (9/9) was attempted to animals attaining the easier criterion before the limit of 30 trials.
Evaluation of Data The avoidance learning and extinction was evaluated in 2 principal ways: (a) by the proportion of animals meeting the criterion in each age group for each gender separately and also for both genders combined; (b) by the number of trials needed to meet criterion in successful animals. Additionally, the following auxiliary parameters were used to analyze the development of avoidance learning: (a) differences between
ONTOGENY OF ACTIVE AVOIDANCE
173
numbers of avoidances within different stages of learning; (b) number of escape reactions during different stages of learning to criterion; (c) number of no reactions during learning to criterion (defined as freezing reactions and/or movements without leaving the electrified platform); (d) number of avoidances and of any stepping on the safe platform within extinguishing until the criterion was met; and (e) comparison of latencies within different stages of learning and extinction. Proportions of animals meeting the criteria were evaluated by means of x2 test; significant differences between the numbers of trials as well as between latencies were estimated by means of Student’s t-test. In retention data a mean proportion of correct responses was evaluated (avoidances in 1-day and I-month avoidance retention, no reactions in retention of extinguished avoidances) within 10 trials. Correct responses in consecutive trials showed the dynamics of retrieval from memory within each age and gender group. Student’s t-test and Kolmogorov-Smirnov 2-sample test were used for the estimation of significant differences.
Results Learning The proportion of animals meeting the less severe criterion of avoidance learning (9/10) for both genders combined increased significantly between the Weeks 2 and 4 (p < .005); the significance of increase between the Weeks 3 and 4 did not reach acceptable levels (p < .05). In our male rats the improvement was steeper between the 2nd and 3rd week whereas in females, between the 3rd and 4th week of life (Fig. 2A). The differences between both genders were significant at the age of 3 weeks (p < ,025). The highest proportion of animals meeting the criterion was attained between the 4th and 6th week of life; this whole period differed significantly (p < .05) from that of 8 weeks and of 3 months, when the proportion was lower (Fig. 2A). An analogous maximum was present with the more severe criterion (9 consecutive correct responses); however the improvement between the 2nd and 4th week was even more evident: 40% in males and 80% in females. The number of trials to the criterion of 90% correct responses decreased steeply between the 2nd and 3rd week, with the lowest values between the Weeks 4 and 6 (Fig. 2B). The differences between the 2nd and all the following weeks were significant in both genders combined and in males, separately; in females they were significant only between Week 2 and Weeks 6 and 8. Because the trials to criterion were calculated with different proportions of animals meeting criterion in males and in females, the obtained curves were biased somewhat. Such a shortcoming did not occur when the number of avoidances during learning was compared in all animals run in the experiment within the 1st and 2nd decade of trials (Fig.2C). Data on meeting the more severe criterion (9/9) yielded practically the same results only the differences were rather less evident due to a smaller number of animals. The optimal learning at the age of 4-6 weeks was even more apparent here, mainly within the 2nd decade. The evaluation of mere escape reactions during learning showed rather ambiguous results with some decrease between Weeks 3 and 8. The only significant difference for
174
MYSLIVECEK AND HASSMANNOVA
both genders combined was between the 2nd and 8th week (Fig. 3A). The comparison of the number of escapes within the 1st and 2nd decade of trials yielded more information, as the decrease of escapes within the 2nd 10 trials (Fig. 3B) was associated with faster avoidance learning mainly between the Weeks 4-6 (see Fig. 2C). The no reactions appeared as another indicator of learning improvement: their number was low in all groups at the age of 4 weeks and more (Fig. 3C). The mean latencies of accomplished avoidances lying between 1.5 and 3.0 sec shortened slightly (by .5 sec, mean) during the development with significant differences between 2 weeks and 3 months in females, and in both genders cornbrned (both p < .01; Fig. 4A). The latencies of the last 5 avoidances were usually, but not significantly, shorter than those of the 1st trials (Fig. 4C). Latencies of escape reactions shortened within learning more reliably and showed also some developmental trends (Fig. 4B).
80-
A 60
-
-. 1. 40-
60
.!
'
-
50 40
30 20
-
- 20 10
20 10
2 3 456 0w3m
2 3 4 5 6 8 ~ 3 m
Fig. 2. (A) Percentage of animals meeting t h e criterion of 9 avoidances in 10 consecutive trials. Abscissa: age. (B) Mean number of trials to criterion 9/10 (according t o the last trial, when the criterion was met). (C) Mean number of avoidances within the 1st (full lines) and 2nd decades of trials (broken line). Significant differences between decades indicated by 1 d o t 0, < .05) or by 2 dots 0, < .01).
ONTOGENY OF ACTIVE AVOIDANCE
U
.b
U
i
175
176
MYSLIVECEK AND HASSMANNOVA A
B
LA
LEsc
1
:1’
-
1
2 3 4 5 6 8 ~ 3rn
2
D
C
I
I
1
3 4 5 6 8w 3m
8
1
’
I
2 3 4 5 6 8w
3m
-
1-1
2 3 4 5 6 8 ~ 3 m
Fig. 4. (A) Mean latency of avoidances to criterion 9/10. (B) Mean latency of escape reaction within avoidance learning to criterion 9/10. (C) Mean latency within the 1st 5 and last 5 avoidances to criterion 9/10. (D) Mean latency of avoidance within extinguishing to the criterion of 9 no reactions in 10 consecutive trials.
Extinction The proportion of animals meeting the 9/10 criterion of extinction within 1 experimental session oscillated slightly during ontogenetic development with a decreasing tendency and an evident fall in 5-week-old animals. These changes were more pronounced when we evaluated the more severe criterion (Fig. SA), by which the differences between the 3rd and 5 t h week were significant in males 0,
Various aspects of the ontogenetic development of learning and memory have been studied on different occasions. Primitive forms of learning appear very early in the ontogeny: in altricial animals they appear practically immediately after birth; in precocial animals simple forms of conditioned reflexes appear before delivery or hatching (Caldwell & Werboff, 1962; Gray, Yates, & McNeal, 1967; Sedlicek, 1963; Thoman, Wetzel, & Levine, 1968; Troshikhin, 1952; for review see Volokhov, 1970). The possibility of conditioning depends on the mutual relations between biological importance (motivational systems) of conditioned and unconditioned stimuli, on their time sequence, and on the maturational characteristics of the functional systems participating in conditioning, as has been already shown (Campbell & Spear, 1972; Myslivecek, 1972). Active 1-way avoidance may be considered as an artificial situation modeling important events of the daily life: the signals and cues guiding this defense reaction need to be learned and remembered: in cats and dogs, for example, escape and active avoidance learning to somesthetic-kinesthetic stimuli has been established before eye-opening (9th day; Bacon, 1973; Bacon & Stanley, 1968; Stanley, Barret, & Bacon,
Reprint requests should be sent to Professor Jaromir Myslivecck, M.D., D.Sc., IHE, Srobarova 48, 10042 Praha 10, Czechoslovakia. Received for publication 14 April 1977 Revised for publication 1 December 1977 Developmental Psychobiology, 12(2): 169-186 (1 979) @ 1979 by John Wiley & Sons, Inc.
0012-1630/79/0012-0169$01.OO
170
MYSLIVECEK AND HASSMANNOVA
1974). Although several papers have dealt with the ontogenetic development of active avoidance in rats covering different periods of life from 19 days (Riccio, Rohrbaugh, & Hodges, 1968) up to 2 years (Dobrovolskaya, 1972; Doty, 1966a,b), no general agreement has been reached as to the ontogeny of active avoidance learning. Some authors deny any changes in learning during ontogeny (Feigley & Spear, 1970; Kirby, 1963; Porter & Thompson, 1967; Riccio et al., 1968). On the other hand, Deneberg and Kline (1958) have described better learning in 60-day rats as compared with 30and 225-day-old animals. Egger and Livesey (1972) have found a monotonically decreasing number of trials to criterion with age in animals at the age of 25, 50, and 100 days. Obraztsova, Dobrovolskaya, Fedorov, and Bogdanova (1972) have: stressed an “exaltation period” of learning and other neural functions between the 1st and 6th month of the rat’s life. Izquierdo, Salzano, ThaddCu, and ThomC (1975) have given an account of better learning in 55-70-day-old rats than in weanlings (20-21 days old) as, similarly, have McLaughlin, Eller, and Korol (1975) in 120-day-old rats compared with 21- and 480-day-old animals. An impairment in old animals is reported also by others (Doty, 1966a,b; Fields, 1953; Freund and Walker, 1971). On the other hand, common agreement prevails in the literature that memory (not only in active avoidance) operates at a higher level in adults than in infant or weanling animals (see Campbell & Spear, 1972; Myslivecek, 1976). Most of the mentioned studies are unfortunately rather episodic, as only some representatives of different developmental periods were investigated. To overcome this gap we have studied the development of avoidance learning and remembering more systematically, starting with the age of 2 weeks up to 3 months (Myslivecek & Hassmannovi, 1973). We have found that optimum 1-way avoidance learning lies between Weeks 4 and 6. The present paper extends and complements our previous work. It deals with different criteria of learning and memory. The aim of the experimentation was to obtain data on active avoidance learning and remembering at precisely delimited ages and to cover with this investigation the lifespan from suckling period (2 weeks) up to adult age ( 3 months). The changes of memory capacity were investigated in 1-day and 1-month retention. The rat pup is an altricial animal with a considerable after-delivery delay before reaching some important degree of early neural and sensory maturation, which happens at the age of about 2 weeks when, among other things, the eyelids open and the acoustic meatuses become unplugged. To avoid developmental shifts at every selected age period the animals had to meet the learning criterion within 1 or 2 experimental days. Thus, we compared animals of increasing age with 1 week intervals up to 6 weeks and then at 8 weeks and at 3 months. A further aim of this study was to find out potential gender-related changes in learning development. This question has an old history (see, for a review of early work, the paper by Tomilin & Stone, 1933). More recently gender differences in avoidance learning (Beatty & Beatty, 1970; Denti & Epstein, 1972; Dobrovolskaya, 1973), development of pyramidal cells in the sensorimotor cortex (Gregory, 1975), maturation of open field behavior (Valley & Boes, 1976), sensitivity to electric shock (Beatty & Fessler, 1976), and ontogeny of self-stimulation behavior (Welley, Stinus, & Cardo, 1975) have been described.
ONTOGENY OF ACTIVE AVOIDANCE
17 1
Methods Materials Albino rats (Rattus norvegicus, Wistar strain) of both genders aged 2,3,4,5,6, and 8 weeks and 3 months, bred in the animal house of the Institute of Physiology, Medical Faculty, Charles University in Prague, were used in the experiment. A total of 146 animals was studied. The animals in each age and gender group (minimum, 10 animals per group) were offspring randomly taken from 4-6 litters.
Apparatus The animals were trained in an automatic 1-way avoidance apparatus described earlier (Myslivecek, Zalud, Dallinger, & Hassmannovi, 1970). Briefly, the apparatus consisted of 2 platforms (one of them electrified) separated during the intertrial intervals (ITI) by a sliding panel. In the present experiment both platforms were flush to equalize the difficulty of experimental task for different ages as far as possible. The strength of the current applied to the electrified floor serving as unconditioned stimulus (US) was adjusted individually for each animal to overreach, just slightly, the threshold levels. The strength of unscrambled footshock which was estimated preliminarily for each age group was changed stepwise within the 1st 2-3 trials, if necessary. Simultaneously with the conditioned stimulus (CS)-a white light on (located on the panel)-the panel was moved to the end of the apparatus and uncovered the safe platform (thus, the CS had a complex character: light and movement of the panel). The C S lasted 15 sec: 5 sec of isolate application, reinforced for another 10 sec by the US. The IT1 duration was 25 sec. The safe platform was accessible during the whole CS and US action and during the 1st 5 sec of IT1 (Fig. 1A). When the program of extinction was begun, the animal was automatically removed by the movement of the panel as soon as it stepped on the safe platform; therefore, we used the term “forced” extinction. The panel then again returned and further access to the safe platform was made possible.
Procedure The experiment started when the animal reached the age of 14,21,. . . days of life, precisely. Of the following 6 experimental tasks all but the last one were studied within 3-4 days (Fig. IB): 1. 1st day-establishment of the avoidance reaction to 1 or both of the 2 criteria to be met (a basic one, 9 correct responses-avoidances-in 10 subsequent trials, and a more severe criterion, 9 consecutive correct responses; only the basic criterion of this 1st task was attempted again the next day if the training on the 1st day was not successful; all other criteria were attempted only once); 2. 2nd (or 3rd) day-a 24-hr retention of the avoidance reaction (10 trials with reinforcement); 3. after 5-min break-forced extinction of avoidance reaction to the same criteria
172
MYSLIVECEK AND HASSMANNOVA
-
A I
cs
:
'
IS'
us
'
18'
:
us
'
MWO 1
cs
' 4
20'
r
:
1
MWO
IS'
'
I
I T 1
I T 1
c----+---t--e---+-
4
B day:
I.
II.
111.
xxx.
Avoidance trammg to criterion
: 4
I
0,
upto 30 trials
Fig. 1. (A) Schedule of CS and US presentation; accessibility of the safe platform initiated and terminated by movements of the sliding wall (MWO) and ITI. (B) Schedule of the experiment.
as with learning (i.e. 90% correct responses within 10 trials and 9 conseculive correct responses); 4. 3rd (or 4th) day-a 24-hr retention of the extinguished avoidance (10 trials without reinforcement); 5. after 5-min break-relearning of avoidance now to a criterion of 75% of correct responses in 20 consecutive trials; 6 . after 4 weeks-a 1-month retention (20 trials with reinforcement). The fulfilment of each criterion had to begin before the 30th trial. If the animal failed to continue with correct performance no further attempts were made to attain the criteria of corresponding task. Only failure at the easier criterion (9/10) stopped further work with the animal. Meeting the more severe criterion (9/9) was attempted to animals attaining the easier criterion before the limit of 30 trials.
Evaluation of Data The avoidance learning and extinction was evaluated in 2 principal ways: (a) by the proportion of animals meeting the criterion in each age group for each gender separately and also for both genders combined; (b) by the number of trials needed to meet criterion in successful animals. Additionally, the following auxiliary parameters were used to analyze the development of avoidance learning: (a) differences between
ONTOGENY OF ACTIVE AVOIDANCE
173
numbers of avoidances within different stages of learning; (b) number of escape reactions during different stages of learning to criterion; (c) number of no reactions during learning to criterion (defined as freezing reactions and/or movements without leaving the electrified platform); (d) number of avoidances and of any stepping on the safe platform within extinguishing until the criterion was met; and (e) comparison of latencies within different stages of learning and extinction. Proportions of animals meeting the criteria were evaluated by means of x2 test; significant differences between the numbers of trials as well as between latencies were estimated by means of Student’s t-test. In retention data a mean proportion of correct responses was evaluated (avoidances in 1-day and I-month avoidance retention, no reactions in retention of extinguished avoidances) within 10 trials. Correct responses in consecutive trials showed the dynamics of retrieval from memory within each age and gender group. Student’s t-test and Kolmogorov-Smirnov 2-sample test were used for the estimation of significant differences.
Results Learning The proportion of animals meeting the less severe criterion of avoidance learning (9/10) for both genders combined increased significantly between the Weeks 2 and 4 (p < .005); the significance of increase between the Weeks 3 and 4 did not reach acceptable levels (p < .05). In our male rats the improvement was steeper between the 2nd and 3rd week whereas in females, between the 3rd and 4th week of life (Fig. 2A). The differences between both genders were significant at the age of 3 weeks (p < ,025). The highest proportion of animals meeting the criterion was attained between the 4th and 6th week of life; this whole period differed significantly (p < .05) from that of 8 weeks and of 3 months, when the proportion was lower (Fig. 2A). An analogous maximum was present with the more severe criterion (9 consecutive correct responses); however the improvement between the 2nd and 4th week was even more evident: 40% in males and 80% in females. The number of trials to the criterion of 90% correct responses decreased steeply between the 2nd and 3rd week, with the lowest values between the Weeks 4 and 6 (Fig. 2B). The differences between the 2nd and all the following weeks were significant in both genders combined and in males, separately; in females they were significant only between Week 2 and Weeks 6 and 8. Because the trials to criterion were calculated with different proportions of animals meeting criterion in males and in females, the obtained curves were biased somewhat. Such a shortcoming did not occur when the number of avoidances during learning was compared in all animals run in the experiment within the 1st and 2nd decade of trials (Fig.2C). Data on meeting the more severe criterion (9/9) yielded practically the same results only the differences were rather less evident due to a smaller number of animals. The optimal learning at the age of 4-6 weeks was even more apparent here, mainly within the 2nd decade. The evaluation of mere escape reactions during learning showed rather ambiguous results with some decrease between Weeks 3 and 8. The only significant difference for
174
MYSLIVECEK AND HASSMANNOVA
both genders combined was between the 2nd and 8th week (Fig. 3A). The comparison of the number of escapes within the 1st and 2nd decade of trials yielded more information, as the decrease of escapes within the 2nd 10 trials (Fig. 3B) was associated with faster avoidance learning mainly between the Weeks 4-6 (see Fig. 2C). The no reactions appeared as another indicator of learning improvement: their number was low in all groups at the age of 4 weeks and more (Fig. 3C). The mean latencies of accomplished avoidances lying between 1.5 and 3.0 sec shortened slightly (by .5 sec, mean) during the development with significant differences between 2 weeks and 3 months in females, and in both genders cornbrned (both p < .01; Fig. 4A). The latencies of the last 5 avoidances were usually, but not significantly, shorter than those of the 1st trials (Fig. 4C). Latencies of escape reactions shortened within learning more reliably and showed also some developmental trends (Fig. 4B).
80-
A 60
-
-. 1. 40-
60
.!
'
-
50 40
30 20
-
- 20 10
20 10
2 3 456 0w3m
2 3 4 5 6 8 ~ 3 m
Fig. 2. (A) Percentage of animals meeting t h e criterion of 9 avoidances in 10 consecutive trials. Abscissa: age. (B) Mean number of trials to criterion 9/10 (according t o the last trial, when the criterion was met). (C) Mean number of avoidances within the 1st (full lines) and 2nd decades of trials (broken line). Significant differences between decades indicated by 1 d o t 0, < .05) or by 2 dots 0, < .01).
ONTOGENY OF ACTIVE AVOIDANCE
U
.b
U
i
175
176
MYSLIVECEK AND HASSMANNOVA A
B
LA
LEsc
1
:1’
-
1
2 3 4 5 6 8 ~ 3rn
2
D
C
I
I
1
3 4 5 6 8w 3m
8
1
’
I
2 3 4 5 6 8w
3m
-
1-1
2 3 4 5 6 8 ~ 3 m
Fig. 4. (A) Mean latency of avoidances to criterion 9/10. (B) Mean latency of escape reaction within avoidance learning to criterion 9/10. (C) Mean latency within the 1st 5 and last 5 avoidances to criterion 9/10. (D) Mean latency of avoidance within extinguishing to the criterion of 9 no reactions in 10 consecutive trials.
Extinction The proportion of animals meeting the 9/10 criterion of extinction within 1 experimental session oscillated slightly during ontogenetic development with a decreasing tendency and an evident fall in 5-week-old animals. These changes were more pronounced when we evaluated the more severe criterion (Fig. SA), by which the differences between the 3rd and 5 t h week were significant in males 0,
