- Email: [email protected]
Acquisition and retention of habituation as a function of intertrial interval duration during training in the blowfly
Behavioural Processes, 15 (1987) 47-57 Elsevier
47
ACQUISITION ANDRETENTION OFHABITUATIONASA FUNCTION OFINTERTRIAL INTERVALDURATIONDURINGTRAININGIN THEBLOWFLY BERNARD THON Centre de Recherche en Biologie du Comportement. UA CNRS664. Universim Paul Sabatier., 118Route de Narbonne. 31062 Toulouse Cedex.France (Accepted
11
May
1987)
ABSTRACT Thon, B., 1987. Acquisition and retention of habituation as a function of intertrial interval duration during training in the blowfly. Be hav . Pro c e s s e s , 1 5 : 4 7 - 5 7 The influence of the duration of the interval between successive stimulations during habituation training on the dynamics of motor response decrement and its retention over time has been studied in the blowfly, &llinhoravomitoria. The intertrial interval (IT11 duration during training was either 7,15, 30.60 or 120 seconds. After a rest period of 30 minutes. a second session was given with the same IT1 duration of 30 seconds for all groups. The response decrement observed during training is faster the longer the IT1 duration. At the end of this session, the lower response level is reached with an IT1 of 60 seconds. After the rest period, the lower responsiveness is observed in the animals which have been trained with the longer ITI. These results are discussed with reference to the dual-process theory of habituation, which can account for the results obtained during training, and memory-centered models to interpret the influence of IT1 duration on retention performance. Key-words: Habituation, Retention, Memory, Intertrial
interval, Insect.
INTRODUCTION The analysis of the relationship
between the frequency
of stimulations and the
rate of response decrement in habituation studies have provided contradictory results, making difficult the linkage between empirical data and the theoretical background. From their review of habituation studies, Thompson and Spencer (1%61 claimed that the more frequent the stimulations, the faster the response decrement. They considered that relation as a general feature of habituation which can be included, among others, in an operational definition
of the phenomenon.
This relation has
been described in a wide variety of animals such as protozoans (Wood. 19701,snails (Cook, 19711.insects (Barras. 19611.fishes (Laming and Ennis. 19821.toads (Brzoska, 19831. frogs (Glanzman and Schmidt, 19791.rats (Askew. 19701or humans (Davis and Henninger,
1972; Gatchel and Lang, 1974; Sane and Ison. 19831. In his pioneer work
on habituation
in the freshwater snail Limnaea staanalis, Pieron (19091 found that
low frequencies of stimulation, i.e. long intertrial intervals (ITIs) increased the number of visual stimulations necessary for the disapearance of the withdrawal response However, habituation was better retained when it has been acquired with an optimum IT1 duration of 10 seconds. With shorter or longer ITIs during previous training, saving on a remote session was reduced, whatever the duration of the rest interval (Pie0376.6357/87/$03.50
Q 1987 Elsevier
Science
Publishers
B.V. (Biomedical
Division)
48
ran, 1911I. Similarly, Davis (197Oa.b) notices that although shorter intervals between acoustic stimul~ions
lead to a faster decrement of the startle response in rats, they
impair long-term retention of habituation. The same results are obtained for the habituation of lick-suppression the rat, or for h~i~~on
(File. 1973) or exploratory bshaviour
of skid-conduc~nce
(Terry, 1979) in
response in humans (Gatchel, 19751.
Moreover. training with distributed (versus massed) trials improves hdituation formance in A&&
per-
(Carew et al., 1972;Carew and Kandel, 19731,the tarentula spider
(Crawford, 19791,the blowfly Cfhon and Pautie, 19841,the Siamese fighting fish (Peeke and Peeke, 1970) o:‘ the white-crowned sparrow (Patterson and Petrinovich, 1979). The influence of intertrial
interval duration on the development of response de-
crement and its retention over time is of co~side~le
interest with respect to con-
current theories of habituation, which can be schematically divided into two groups: 1) those assuming that response decrement comes from a self-generated synaptic depression located in the underlying
pathway from sensory input to motor ouput
(Thompson and Spencer, 1%6; Horn. 1967; Groves and Thompson, 1970: Petrinovich, 19841and 21 those assuming that response waning results from an active inhibitory process coming from collateral structures, extrinsic to the S-R pathway (Sokolov, 1963;Stein, 1%6; Wagner, 19761979; Whitlow and Wagner, 1984). Generally, the advocates of the former conception are primarly interested in the acquisition phase of h~~~~on,
and predict that the shorter the ITI, the faster (or deeper) the response
decrement. On the other hand, the supporters of the latter are rather concerned with the memory phase of habituation, i.e. the long-term retention of the response decrement. and assume that long ITI will facilitate inform~on
processing (rehearsal, as-
sociation between the releasing stimulus and contextual cues ...I which will lead to a better retention of habituation over time (Wagner, 197619791. The only available data dealing with this problem come from vertebrate studies (Davis, 1970a, b: File, 1973; Gatchel. 1975;Terry, 1979). and we were interested to analyse the influence of ITI duration on habituation in an insect, to provide some comparative data to this field of research. This work has been conduced in the bowfly &I&& ~~
which has been previously used in habituation studies (Than and
Queinnec, 1976;Thon, 1980;Thon and Pauzie, 1984), We have analysed the influence of the duration of the interval between successive visual emulsions
on the h~ituatio~
of the optomotor response. As underlined by Davis (19701,each stimulation during an habituation session is both a training animal So, in order to avoid confoun~ng
trial and a test of the responsiveness
of the
ITI duration as a test interval and a training
interval, the animals have been trained with different IT1 duration, and tested in a remote session in which IT1 duration is identical for all experimental groups (Davis, 1970;Gatchei, 1974).
49
All experiments have been conducted in adult blowflies m ~Q&Q& of both sexes, 3-7 days: old. They have been reared in the laboratory from the last larval instar. The method of recording motor activity has been previously described (Than, 1982; Thon and Pautie, 1984).The fly is tethered by the thorax to the tip of a needle and placed above two brass grids separated by a two-millimeters space. The legs of the animal are in conctact with the left and right grids. One grid is connected to an input of an amplifier pen-recorder, while the other grid is connected to the positive pole of a battery (I .5volts). The negative pole of the battery is connected to the other plug of the input line of the amplifier. Each movement of the fly’s legs induces a change in the overall resistance of this circuit and the variations of the weak current intensity are amplified and recorded on paper. The fly being in darkness, the visual stimulation is given by lightening a rotating striped disk (35 cm. in diameter) with 18 alternating white and black sectors, placed at 12 cm beside the fly. The rotating speed of the striped disk is 360 degrees/second. A photocell detects the onset and offset of the stimulation which are recorded parallely to motor activity on polygrah paper. In the whole experiment, the visual stimulation has a constant duration of 2 seconds. Before experiment, each fly is randomly assigned to one of five experimental groups, composed of 10 males and 10 females. Each group receives a first session of 20 visual stimulations, with different intertrial interval durations. These durations are 7 sec., 15 sec., 30 sec., 60 sec. and 120 sec. It must be noticed that if a fly does not respond to any one of the first three stimulations, it is discarded from the experiment. After the 20th stimulation, the fly is left in darkness in the experimental apparatus for a rest period of 30 minutes. A second series of 30 stimulations is then given, which constitutes the retention test session. During this session, the intertrial interval duration is equal to 30 sec. for all groups. RESULTS Motor responses are evaluated in an all-or-none
fashion, since their amplitude is
difficult to be estimated. We consider that a motor response has been released if legs movements have been recorded during stimulus presentation
or within the two se-
conds following stimulus offset. The Figure 1 represents the percent of evoked motor responses for each stimulation of the two sessions, During training, all groups except the ITI: group, show a significant decrease of the number of released responses throughout stimulus repetition Kochran 0 test, p <.OlI. In order to allow a between group comparison, animals performance has been scored as the number of responses displayed by five trials blocks. A 2x5x4 three-ways analysis of variance (Wirier, 1971) with sex and IT1 duration as independent factors and blocks of trials as dependent factor, shows no differences between males and females (F ( 1). The duration of IT1 has a strong effect ( F (4.90) = 11.21,p <,001I as well as blocks of trials ( F (3.270) = 33.75, p < ,001).The significant
interaction between ITI
duration and blocks of trials ( F (12,270) = 2.35, p ( ,011indicates that response decrement proceeds at different speeds in function of ITI duration.
50
IOOZ, a0 60 70 60
so 40 20 IO
I-
____________________ 111:
7
300 k p.."...;"_....', 1002 1 IO so 00 70 60 so ________--____-40 10 20 111: 1s IO I0 lOOZ,_ I 10
20
_
-
L+.-."..;....""',
20
t”“““:““““‘:“‘“““, 1
10
20
30
I
IO
20
30
t”“““:““““‘:““““L( I
IO
20
30
10
20
30
so
60 70 60 so -_--___--__ __ _-40 30 1% 20 111: 30 10 t 0 IOOZ 1 20 SD SD 70 %I 60 so _--__ ___ ___--___-40 30 20 171: 60 10 0 100): 1 20 10 SO SO 70 60 so _--_ __-_ _ _--__-40 30 20 111: 120 IO IL 0 1 20 10
L,.......;......,..,
10
Lh L,........;.........,
t”“““:““““‘;““““‘~
~“““‘;“‘-“:,““.“‘,
1
Figure 1: Variation of the percent of responding animals (ordinates) during training and during retention test in the different ITI groups. Abscissa stimulation number, The dashed line indicates the 50% level The duration of the rest period is 30 minutes, To quantify the influence of ITI duration on response decrement, two parameters can be used: 11the number of stimulations necessary to reach a given response level and 2) the response level reached after a given number of stimulations. The number of stimulations necessary to suppress the motor response in more than 50% of the animals is greater than 20 in the ITI: group and is equal to 17,12,10 and 6 respectively for ITI:15. ITI:30, IT150 and ITI:
groups. The level of response reached at the end
51
of the session is respectively equal to 55%. 35%. 30%. 10% and 35%. So, initial rate of response decrement appears to be faster the longer the IT1 duration, but, in terms of final response level, response decrement is more pronounced for an intermediate IT1 duration of 60 sec., which constitutes, in the present experiment, an optimum interval for the acquisition of behavioural habituation. After the rest period of 30 minutes, the first visual stimulation of the second session induces a response in 65%, 55%. 60%. 20%. and 10% of the animals of ITI:7, IT1:15, ITI:30, 1TI:COand ITI:
groups respectively. This difference is significant
(Chi-
square test, p <,011.The responsiveness of the animals after the rest period appears to be even lower that the IT1 duration was longer during previous training.
It can be
noticed that spontaneous recovery during the rest period is not important, with respect to the final response level at the end of training.
Moreover, for the ITI:
group, response level is lower after the rest period than before, which can be seen as an improvement of habituation when stimulations are stopped. Apart from this group, it appears a good agreement between response level after and before the rest period, and we can suspect that the retention performance
is directly related to acquisition
performance at the individual level. To test this hypothesis, we have computed the biserial correlation coeficient between the number of diplayed responses during the last five stimulations of training on one hand, and the presence or absence of response to the first visual stimulation of the retention
session on the other hand. The larger value of rbis is .30 for the ITI:
group, and it is not significant.
So, the final response level at the end of training
is
not a predictor of the responsiveness of the animals after the rest period. Another way to analyse the relationship
between acquisition and retention perfor-
mance is to split each experimental group into subgroups which differ according to the final response level or to their responding to the first stimulation of retention test. Figure 2 shows the response curves computed in animals which have reached a low level of responsiveness at the end of training
(less than 2 responses for the last 5
stimulations: T- 1 and those which display a high response level (more than 3 responses: T+ 1.This treatment cannot be applied to the ITI:
group in which the size of the
T+ subgroup is too small. During training, T- and T+ curves just diverge during the last four stimulations. Within each IT1 group, the two subgroups show the same response level to the first stimulation of retention test, which is another demonstration of the independence
of the responsiveness
of the animals before and after the rest
period. When stimulations are repeated again in the second session, the T+ animals tend to display a higher number of responses than the T- ones, which suggests that sustained responsiveness ristic.
during stimulus iterations could be an individual characte-
Figure 2: Response curves of animals which display less than 2 responses (dashed line) or more than 3 responses (solid line) during the last five trials of training session. The site of the subgroups is indicated in the right of each graph.
Similarly, we can divide the experimental groups according to the presence or absence of response to the first stimulation following the rest period. Figure 3 represents the response curves of the responding (R+1and no responding (R-j animals to that stimulation. The IT160 and ITI:
groups are not included since the number of
R* animals is too small in these conditions. In the other IT1 groups, R+ and R- animals show the same pattern of response variation during training. So, the release of a response to the first stimulation of retention test cannot be ascribed to a faster or deeper
53
response decrement during the training
session. When stimulations are repeated
again, R+ animals display a regular response decrement, while R-animals show an initial rise up of response level, in such a way that the two curves join each other for the 7th stimulation. This phenomenon is accurately reproduced in the ITI:7, ITI: ITI:
and
gMUpS.
1
10
20
30
1OOl so SO 70 60 60 40 30 20 10 0
_I
IO
20
-
30
Figure 3: Response curves in animals which do not respond (dashed line) and animals which respond (solid line) to the first stimulation of the retention session after the rest period.
DISCUSSION The present results reveal a clear-cut effect of intertrial
interval duration on the
dynamic of response variation in the course of habituation training. For the shortest IT1 (7 sec.). no evidence for behavioural habituation can be found, since the response level remains steady for the entire session. When ITI durations becomes longer. a decline in the number of evoked responses occurs with stimulus repetition. The initial
54
rate of response decrement is directly related to the length of the ITI. Such a result is not so often encountered in the literature where an inverse relationship is generally reported. The dual-process theory of Groves and Thompson (1970) suggests some lines of interpretation which could resume that apparent contradiction. That theory assumes that stimulus repetition induces the development of two antagonistic
processes:
each stimulation is supposed both to depress the response system (habituation) and to raise up the general reactivity state of the animal (sensitization). The response amplitude (or probability) to a given trial will be a function of the actual balance between these two processes. Experimental evidences (Groves et al., 1969; Groves and Thompson, 1970; Davis, 1972; Thompson et al., 1973; Petrinovich,
1984) suggest that they are
independent and proceed with different time constants: once initiated, sensitixation decays faster over time than habituation. So, we may suggest that when IT1 duration is short enough, each stimulation occurs while the sensitizing one still persists, preventing the IT1 is lengthened,
effects of the preceding
a decrease in the actual response. When the duration of
each stimulation intervenes
at a time when sensitixing
in-
fluences of the previous one has decayed, which will allows the habituation process to be revealed by a waning
of the response. However, if IT1 duration becomes long
enough, it leads to a disapearance of sensitizing influences
but also to a vanishing
of
habituation (partial recovery) and response decrement will be less pronounced. This would explain why an optimum IT1 duration (60 sec. in the present experiment) produces a deeper response decrement during training than shorter or longer ones. However, if the dual-process theory can appropriately account for the observed results during habituation training, the data dealing with retention performance after the rest period can hardly be explained within this theoretical framework. The best measure of habituation retention is given by the responsiveness
of the animals to the
first stimulation of the retention session, since sensitixing effects are not yet triggered by stimulus repetition. The retention of habituation appears to be better when the IT1duration was longer during previous training. Three hypotheses could be suggested to account for this result: I I a residual sensitization. the level of which is higher the shorter the ITI, still remains after the rest period. 2) short ITIs impair the development of habituation process during training. 3) even if habituation took place during training, short ITIs prevent its long-term retention. The first two hypothesis imply that a correlation could be found between training performance, evaluated by the final response level. and retention score, estimated by response likelihood after the rest period. Such a correlation cannot be found, since it is impossible to predict if an animal will respond to the visual stimulation following the rest period from its response level at the end of training (Fig. 2). Moreover, ani-
55
mals which respond and those which do not respond to that stimulation display exactly the same pattern of response variation during training
session (Fig. 31. This lack
of correlation between acquisition and retention performance has been often reported (Then. 1981;Thon and Pat&, 19841and suggests an independence of the processes underlying
short-term and long-term retention of habituation. This independence is
also supported by comparing the scores of ITI:
and ITI
groups: while these
groups reach the same response level at the end of training, response likelihood after the rest period raises up in the ITI: suggesting some reminiscence-like
animals and falls down in the ITI
group,
process in the latter. These arguments support
the first hypothesis which postulates a differential effect of ITI duration on shortterm and long-term retention of habituation.
Davis (1970aI suggests that short ITIs
prevent the consolidation of habituation in long-term memory: habituation, triggered by a given stimulation would be disrupted if the next stimulation is applied after a short interval. Other experimental data show that response decrement observed with high frequency
of stimulation results from refractory-like
processes (Wilson and
Groves, 1973) which could impair the development of long-term habituation. According to Wagner (1976,19791, a lack of responding to a stimulus results from a priming of the stimulus representation
into short-term memory storage. This pri-
ming can be induced by a recent occurence of that stimulus (self-generated priming) or by the presentation of environmental
cues previously associated with the stimulus
during initial acquisition session (retrieved-generated
priming). Self-generated pri-
ming is involved in short-term retention of habituation and plays an important part in response decrement during training, while retrieved-generated
priming is a de-
terminant of long-term retention of habituation (Whitlow, 19751.The development of the association between contextual cues and the to-be-habituated stimulus would be improved by long intervals between successive trials. Even if the presents results can hardly be accounted for by this theoretical model, since the animals were only presented with the releasing stimulus with a few -if any- environmental
cues, they sug-
gest that behavioural habituation in the blowfly involves several independent cesses during acquisition and retention, and that memory-centered to be restricted to vertebrates, but could be generalized to invertebrate
pro-
models have not habituation.
Another point which deserves attention in the present results, is the tendency for a transient rise in response level during the second session by animals which are not responsive at the end of the rest period. Such a phenomenon
has been previously
described in the blowfly (Than and Pauzie, 19841,and seems not to be an artifact. It appears to be independent of IT1 duration (Fig. 3). It can be considered as the result of the development of sensitization when the stimulation is repeated again. We can presume that this transient sensitization cannot be revealed in responding animals because of ceiling effects. No explanation of this phenomenon
can be proposed, but it
56
stresses the importance of the analysis of individual pattern of response variation in habituation studies, which reveales interesting
features generally hiden when the
data are averaged within experimental groups. REFERENCES Askew, HR, 1970. Effects of stimulus intensity and intertrial interval on habituation of the head-shake response in the rat. J. Camp. Physiol. PsychoI.. 72: 492497, Barras, R., 1961. A quantitative study of the behaviour of the male Mormoniella u (Hym. pteromalidae) towards two constant stimuli situations, Be2 viour, 18:288312. Brzoska, J., 1983. Amplitude, latency and habituation of the electrodermal response to acoustic stimuli in the frog. Behav. Proc., 8: 229-242. Carew. T.J. and Kandel, E.R.. 1973. Acquisition and retention of long-term habituation in Aplysia: correlation of behavioral and cellular processes. Science. 182: 1158-1160. Carew, T.J., Pinsker, H.M.and Kandel, E.R.1972.Long-term habituation of a defensive withdrawal1 reflex in At,&&. Science, 175.451-454. Cook, A., 1971. Habituation in the freshwater snail tm MI. Anim. Behav., 19:463-474. Crawford, F.T., 1979.The effect of distribution of trials upon the habituation of tonic immobility in the tarentula mm. Bull. Psychon. Sot., 14: 135-137. Davis, M., 1970a. Effect of interstimulus length and variability on startle response habituation in the rat. J. Camp. Physiol. Psychol., 72: 177-192. Davis, M, 1970b. Interstimulus interval and startle response habituation with a ‘control’ for total time during training. Psychon. Sci.. 20: 39-41. Davis, M., 1972.Differential retention of sensitization and habituation of the startle response in the rat. J. Camp. Physiol. Psychol., 78: 260-267. Davis M. and Henninger, G.R..1972.Comparison of response plasticity between the eyeblink and vertex potential in humans. Electoencephal. Clin. Neurophysiol., 33:283-293. File, SE., 1973. Interstimulus interval and the rate of behavioural habituation. Quart. J. Exp. Psychol., 25: 360-367. Gatchel, R.J., 1975.Effects of interstimulus interval length on short- and long-term habituation of autonomic components of the orienting response. Physiol. Psychol., 3: 133-136. Gatchel, R.J. and Lang, P.J., 1974. Effects of interstimulus interval length and variability on habituation of autonomic components of the orienting response. J. Exp. Psychol., 103: 802-804. GIantman. D.L.and Schmidt, EC., 1979.Habituation of the nictitating membrane reflex in the intact frog. Physiol. Behav.. 22: 1141-1148. Groves, P.M., Lee, D.,and Thompson, R.F., 1969.Effects of stimulus frequency and intensity on habituation and sensitization in acute spinal cat. Physiol. Behav.. 4: 383-388. Groves, P.M. and Thompson, R.F.. 1970. Habituation: a dual-process theory Psychol. Rev., 77: 419450. Horn, G., 1%7. Neuronal mechanisms of habituation. Nature, 215: 707-711. Laming, P.R. and Ennis, P., 1982. Habituation of fright and arousal in the teleosts Csrassiusmand&&&&ifhs J. Camp. Physiol. Psychol., 96: 460-466. Patterson, T.L. and Petrinovich. L., 1979.Field studies of habituation. II) Effects of massed stimulus presentation. J. Camp. Physiol. Psychol., 93: 351-359. Peeke, H.V.S.and Peeke, SC, 1970. Habituation of conspecific aggressive responses in the Siamese fighting fish &&&solende&. Behaviour. 36: 232-245. Petrinovich, L., 1984.A two-factor dual-process theory of habituation and sensitiza-
57
tion. In H.V.S.Peeke and L. Petrinovich, (Edititorsl, Habituation, Sensitization and Behavior, Academic Press, New York, pp 17-55. Pieron. II., 1909. L’adaptation aux obscurations repetees comme phenomene de memoire chez les animaux inferieurs. Arch.Psychol., 9: 39-50. Pieron. II.. 1911.Sur la determination de la periode d’etablissement dans les acquisitions mnemoniaues. C.R. Acad. Sci. Paris, 152: 1410-1411. Sanes, J.N. and Ison,-J.R., 1983. Habituation and-sensitixation of components of the human eyeblink reflex. Behav. Neurosci., 97: 833-836. Sokolov. E.N., 1%3. Perception and the conditional reflex. Pergamon Press, New York. Stein, L., 1966. Habituation and stimulus novelty: a model based on classical conditionning. Psychol. Rev., 73: 352356. Terry, W.S., 1979. Habituation and dishabituation of rat’s exploration of a novel environment. Anim. Learn. Behav.. 7: 525-536. Thompson, R.F.. Groves, P.M., Teyler, T.J. and Roemer, R.A., 1973. A dual-process theory of habituation: theory and behavior. In H.V.S.Peeke and M.J. Hen (Editors). Habituation. Vol.1. Behavioral Studies. Academic press, New York, pp 239-271. Thompson R.F. and Spencer, W.A., 1%6. Habituation: a model phenomenon for the study of neuronal substrates of behavior. Psychol. Rev., 173: 16-43. Thon, B.. 1980.Habituation of cardiac and motor responses to a moving visual stimulus in the blowfly. J. Comp. Physiol. Psychol., 94: 886-893. Thon, B.. 1981. Etude quantitative de la retention de l’habituation de la reponse optocardiaque chet Calliohora vomitoria. Physiol. Behav., 26: 423-432. Thon, B.. 1982. Influence of the cardiac phase on the latency of a motor response to avisual stimulus in the blowfly. J. Insect Physiol., 28: 411-416. Thon, B. and Pauzie, A., 1984. Differential sensitization, retention and generalization of habituation in two response systems in the blowfly (Calliohoravomitod). J. Comp.Psychol., 98: 119-130. Thon, B. and Queinnec, Y., 1976.Habituation de la reponse optocardiaque chex calli_ W~Q&Q&. Biol. Behav., 2: 101-113. Wagner, A.R.. 1976. Priming in STM: an information processing mechanism for self-generated and retrieved-generated depression in performance. In: T.J. Tighe and N. Leaton (Editors). Habituation: perspectives from child development, animal behavior and neurophysiology. Lawrence Erlbaum Ass.. Hillsdale, pp. 95-128. Wagner, A.R., 1979. Habituation and memoryIn: A. Dickinson and R.A. Boakes (Editors). Mechanisms of learning and motivation. A memorial volume for J. Konorski. Lawrence Erlbaum Ass., Hillsdale, pp. 53-82. Whitlow, J.W.. 1975. Short-term memory in habituation and dishabituation. J. Exp. Psychol. Anim. Behav. Proc., 104: 189-206. Whitlow, J.W. and Wagner, A.R., 1984. Memory and habituation. In H.V.S.Peeks and L. Petrinovich, (Editors), Habituation, Sensitixation and Behavior. Academic Press, New York, pp. 103-153. Wilson, C.J. and Groves, P.M., 1973. Refractory period and habituation of acoustic startle response in the rat. J. Comp. Physiol. Psychol., 83: 492-498. Wirier. J., 1971. Statistical principles in experimental design. McGraw Hill, New York. Wood,DC., 1970.Parametric study of the response decrement produced by mechanical stimuli in the protozoan Stentor coeruleus J. Neurobiol., 1: 345-360.
47
ACQUISITION ANDRETENTION OFHABITUATIONASA FUNCTION OFINTERTRIAL INTERVALDURATIONDURINGTRAININGIN THEBLOWFLY BERNARD THON Centre de Recherche en Biologie du Comportement. UA CNRS664. Universim Paul Sabatier., 118Route de Narbonne. 31062 Toulouse Cedex.France (Accepted
11
May
1987)
ABSTRACT Thon, B., 1987. Acquisition and retention of habituation as a function of intertrial interval duration during training in the blowfly. Be hav . Pro c e s s e s , 1 5 : 4 7 - 5 7 The influence of the duration of the interval between successive stimulations during habituation training on the dynamics of motor response decrement and its retention over time has been studied in the blowfly, &llinhoravomitoria. The intertrial interval (IT11 duration during training was either 7,15, 30.60 or 120 seconds. After a rest period of 30 minutes. a second session was given with the same IT1 duration of 30 seconds for all groups. The response decrement observed during training is faster the longer the IT1 duration. At the end of this session, the lower response level is reached with an IT1 of 60 seconds. After the rest period, the lower responsiveness is observed in the animals which have been trained with the longer ITI. These results are discussed with reference to the dual-process theory of habituation, which can account for the results obtained during training, and memory-centered models to interpret the influence of IT1 duration on retention performance. Key-words: Habituation, Retention, Memory, Intertrial
interval, Insect.
INTRODUCTION The analysis of the relationship
between the frequency
of stimulations and the
rate of response decrement in habituation studies have provided contradictory results, making difficult the linkage between empirical data and the theoretical background. From their review of habituation studies, Thompson and Spencer (1%61 claimed that the more frequent the stimulations, the faster the response decrement. They considered that relation as a general feature of habituation which can be included, among others, in an operational definition
of the phenomenon.
This relation has
been described in a wide variety of animals such as protozoans (Wood. 19701,snails (Cook, 19711.insects (Barras. 19611.fishes (Laming and Ennis. 19821.toads (Brzoska, 19831. frogs (Glanzman and Schmidt, 19791.rats (Askew. 19701or humans (Davis and Henninger,
1972; Gatchel and Lang, 1974; Sane and Ison. 19831. In his pioneer work
on habituation
in the freshwater snail Limnaea staanalis, Pieron (19091 found that
low frequencies of stimulation, i.e. long intertrial intervals (ITIs) increased the number of visual stimulations necessary for the disapearance of the withdrawal response However, habituation was better retained when it has been acquired with an optimum IT1 duration of 10 seconds. With shorter or longer ITIs during previous training, saving on a remote session was reduced, whatever the duration of the rest interval (Pie0376.6357/87/$03.50
Q 1987 Elsevier
Science
Publishers
B.V. (Biomedical
Division)
48
ran, 1911I. Similarly, Davis (197Oa.b) notices that although shorter intervals between acoustic stimul~ions
lead to a faster decrement of the startle response in rats, they
impair long-term retention of habituation. The same results are obtained for the habituation of lick-suppression the rat, or for h~i~~on
(File. 1973) or exploratory bshaviour
of skid-conduc~nce
(Terry, 1979) in
response in humans (Gatchel, 19751.
Moreover. training with distributed (versus massed) trials improves hdituation formance in A&&
per-
(Carew et al., 1972;Carew and Kandel, 19731,the tarentula spider
(Crawford, 19791,the blowfly Cfhon and Pautie, 19841,the Siamese fighting fish (Peeke and Peeke, 1970) o:‘ the white-crowned sparrow (Patterson and Petrinovich, 1979). The influence of intertrial
interval duration on the development of response de-
crement and its retention over time is of co~side~le
interest with respect to con-
current theories of habituation, which can be schematically divided into two groups: 1) those assuming that response decrement comes from a self-generated synaptic depression located in the underlying
pathway from sensory input to motor ouput
(Thompson and Spencer, 1%6; Horn. 1967; Groves and Thompson, 1970: Petrinovich, 19841and 21 those assuming that response waning results from an active inhibitory process coming from collateral structures, extrinsic to the S-R pathway (Sokolov, 1963;Stein, 1%6; Wagner, 19761979; Whitlow and Wagner, 1984). Generally, the advocates of the former conception are primarly interested in the acquisition phase of h~~~~on,
and predict that the shorter the ITI, the faster (or deeper) the response
decrement. On the other hand, the supporters of the latter are rather concerned with the memory phase of habituation, i.e. the long-term retention of the response decrement. and assume that long ITI will facilitate inform~on
processing (rehearsal, as-
sociation between the releasing stimulus and contextual cues ...I which will lead to a better retention of habituation over time (Wagner, 197619791. The only available data dealing with this problem come from vertebrate studies (Davis, 1970a, b: File, 1973; Gatchel. 1975;Terry, 1979). and we were interested to analyse the influence of ITI duration on habituation in an insect, to provide some comparative data to this field of research. This work has been conduced in the bowfly &I&& ~~
which has been previously used in habituation studies (Than and
Queinnec, 1976;Thon, 1980;Thon and Pauzie, 1984), We have analysed the influence of the duration of the interval between successive visual emulsions
on the h~ituatio~
of the optomotor response. As underlined by Davis (19701,each stimulation during an habituation session is both a training animal So, in order to avoid confoun~ng
trial and a test of the responsiveness
of the
ITI duration as a test interval and a training
interval, the animals have been trained with different IT1 duration, and tested in a remote session in which IT1 duration is identical for all experimental groups (Davis, 1970;Gatchei, 1974).
49
All experiments have been conducted in adult blowflies m ~Q&Q& of both sexes, 3-7 days: old. They have been reared in the laboratory from the last larval instar. The method of recording motor activity has been previously described (Than, 1982; Thon and Pautie, 1984).The fly is tethered by the thorax to the tip of a needle and placed above two brass grids separated by a two-millimeters space. The legs of the animal are in conctact with the left and right grids. One grid is connected to an input of an amplifier pen-recorder, while the other grid is connected to the positive pole of a battery (I .5volts). The negative pole of the battery is connected to the other plug of the input line of the amplifier. Each movement of the fly’s legs induces a change in the overall resistance of this circuit and the variations of the weak current intensity are amplified and recorded on paper. The fly being in darkness, the visual stimulation is given by lightening a rotating striped disk (35 cm. in diameter) with 18 alternating white and black sectors, placed at 12 cm beside the fly. The rotating speed of the striped disk is 360 degrees/second. A photocell detects the onset and offset of the stimulation which are recorded parallely to motor activity on polygrah paper. In the whole experiment, the visual stimulation has a constant duration of 2 seconds. Before experiment, each fly is randomly assigned to one of five experimental groups, composed of 10 males and 10 females. Each group receives a first session of 20 visual stimulations, with different intertrial interval durations. These durations are 7 sec., 15 sec., 30 sec., 60 sec. and 120 sec. It must be noticed that if a fly does not respond to any one of the first three stimulations, it is discarded from the experiment. After the 20th stimulation, the fly is left in darkness in the experimental apparatus for a rest period of 30 minutes. A second series of 30 stimulations is then given, which constitutes the retention test session. During this session, the intertrial interval duration is equal to 30 sec. for all groups. RESULTS Motor responses are evaluated in an all-or-none
fashion, since their amplitude is
difficult to be estimated. We consider that a motor response has been released if legs movements have been recorded during stimulus presentation
or within the two se-
conds following stimulus offset. The Figure 1 represents the percent of evoked motor responses for each stimulation of the two sessions, During training, all groups except the ITI: group, show a significant decrease of the number of released responses throughout stimulus repetition Kochran 0 test, p <.OlI. In order to allow a between group comparison, animals performance has been scored as the number of responses displayed by five trials blocks. A 2x5x4 three-ways analysis of variance (Wirier, 1971) with sex and IT1 duration as independent factors and blocks of trials as dependent factor, shows no differences between males and females (F ( 1). The duration of IT1 has a strong effect ( F (4.90) = 11.21,p <,001I as well as blocks of trials ( F (3.270) = 33.75, p < ,001).The significant
interaction between ITI
duration and blocks of trials ( F (12,270) = 2.35, p ( ,011indicates that response decrement proceeds at different speeds in function of ITI duration.
50
IOOZ, a0 60 70 60
so 40 20 IO
I-
____________________ 111:
7
300 k p.."...;"_....', 1002 1 IO so 00 70 60 so ________--____-40 10 20 111: 1s IO I0 lOOZ,_ I 10
20
_
-
L+.-."..;....""',
20
t”“““:““““‘:“‘“““, 1
10
20
30
I
IO
20
30
t”“““:““““‘:““““L( I
IO
20
30
10
20
30
so
60 70 60 so -_--___--__ __ _-40 30 1% 20 111: 30 10 t 0 IOOZ 1 20 SD SD 70 %I 60 so _--__ ___ ___--___-40 30 20 171: 60 10 0 100): 1 20 10 SO SO 70 60 so _--_ __-_ _ _--__-40 30 20 111: 120 IO IL 0 1 20 10
L,.......;......,..,
10
Lh L,........;.........,
t”“““:““““‘;““““‘~
~“““‘;“‘-“:,““.“‘,
1
Figure 1: Variation of the percent of responding animals (ordinates) during training and during retention test in the different ITI groups. Abscissa stimulation number, The dashed line indicates the 50% level The duration of the rest period is 30 minutes, To quantify the influence of ITI duration on response decrement, two parameters can be used: 11the number of stimulations necessary to reach a given response level and 2) the response level reached after a given number of stimulations. The number of stimulations necessary to suppress the motor response in more than 50% of the animals is greater than 20 in the ITI: group and is equal to 17,12,10 and 6 respectively for ITI:15. ITI:30, IT150 and ITI:
groups. The level of response reached at the end
51
of the session is respectively equal to 55%. 35%. 30%. 10% and 35%. So, initial rate of response decrement appears to be faster the longer the IT1 duration, but, in terms of final response level, response decrement is more pronounced for an intermediate IT1 duration of 60 sec., which constitutes, in the present experiment, an optimum interval for the acquisition of behavioural habituation. After the rest period of 30 minutes, the first visual stimulation of the second session induces a response in 65%, 55%. 60%. 20%. and 10% of the animals of ITI:7, IT1:15, ITI:30, 1TI:COand ITI:
groups respectively. This difference is significant
(Chi-
square test, p <,011.The responsiveness of the animals after the rest period appears to be even lower that the IT1 duration was longer during previous training.
It can be
noticed that spontaneous recovery during the rest period is not important, with respect to the final response level at the end of training.
Moreover, for the ITI:
group, response level is lower after the rest period than before, which can be seen as an improvement of habituation when stimulations are stopped. Apart from this group, it appears a good agreement between response level after and before the rest period, and we can suspect that the retention performance
is directly related to acquisition
performance at the individual level. To test this hypothesis, we have computed the biserial correlation coeficient between the number of diplayed responses during the last five stimulations of training on one hand, and the presence or absence of response to the first visual stimulation of the retention
session on the other hand. The larger value of rbis is .30 for the ITI:
group, and it is not significant.
So, the final response level at the end of training
is
not a predictor of the responsiveness of the animals after the rest period. Another way to analyse the relationship
between acquisition and retention perfor-
mance is to split each experimental group into subgroups which differ according to the final response level or to their responding to the first stimulation of retention test. Figure 2 shows the response curves computed in animals which have reached a low level of responsiveness at the end of training
(less than 2 responses for the last 5
stimulations: T- 1 and those which display a high response level (more than 3 responses: T+ 1.This treatment cannot be applied to the ITI:
group in which the size of the
T+ subgroup is too small. During training, T- and T+ curves just diverge during the last four stimulations. Within each IT1 group, the two subgroups show the same response level to the first stimulation of retention test, which is another demonstration of the independence
of the responsiveness
of the animals before and after the rest
period. When stimulations are repeated again in the second session, the T+ animals tend to display a higher number of responses than the T- ones, which suggests that sustained responsiveness ristic.
during stimulus iterations could be an individual characte-
Figure 2: Response curves of animals which display less than 2 responses (dashed line) or more than 3 responses (solid line) during the last five trials of training session. The site of the subgroups is indicated in the right of each graph.
Similarly, we can divide the experimental groups according to the presence or absence of response to the first stimulation following the rest period. Figure 3 represents the response curves of the responding (R+1and no responding (R-j animals to that stimulation. The IT160 and ITI:
groups are not included since the number of
R* animals is too small in these conditions. In the other IT1 groups, R+ and R- animals show the same pattern of response variation during training. So, the release of a response to the first stimulation of retention test cannot be ascribed to a faster or deeper
53
response decrement during the training
session. When stimulations are repeated
again, R+ animals display a regular response decrement, while R-animals show an initial rise up of response level, in such a way that the two curves join each other for the 7th stimulation. This phenomenon is accurately reproduced in the ITI:7, ITI: ITI:
and
gMUpS.
1
10
20
30
1OOl so SO 70 60 60 40 30 20 10 0
_I
IO
20
-
30
Figure 3: Response curves in animals which do not respond (dashed line) and animals which respond (solid line) to the first stimulation of the retention session after the rest period.
DISCUSSION The present results reveal a clear-cut effect of intertrial
interval duration on the
dynamic of response variation in the course of habituation training. For the shortest IT1 (7 sec.). no evidence for behavioural habituation can be found, since the response level remains steady for the entire session. When ITI durations becomes longer. a decline in the number of evoked responses occurs with stimulus repetition. The initial
54
rate of response decrement is directly related to the length of the ITI. Such a result is not so often encountered in the literature where an inverse relationship is generally reported. The dual-process theory of Groves and Thompson (1970) suggests some lines of interpretation which could resume that apparent contradiction. That theory assumes that stimulus repetition induces the development of two antagonistic
processes:
each stimulation is supposed both to depress the response system (habituation) and to raise up the general reactivity state of the animal (sensitization). The response amplitude (or probability) to a given trial will be a function of the actual balance between these two processes. Experimental evidences (Groves et al., 1969; Groves and Thompson, 1970; Davis, 1972; Thompson et al., 1973; Petrinovich,
1984) suggest that they are
independent and proceed with different time constants: once initiated, sensitixation decays faster over time than habituation. So, we may suggest that when IT1 duration is short enough, each stimulation occurs while the sensitizing one still persists, preventing the IT1 is lengthened,
effects of the preceding
a decrease in the actual response. When the duration of
each stimulation intervenes
at a time when sensitixing
in-
fluences of the previous one has decayed, which will allows the habituation process to be revealed by a waning
of the response. However, if IT1 duration becomes long
enough, it leads to a disapearance of sensitizing influences
but also to a vanishing
of
habituation (partial recovery) and response decrement will be less pronounced. This would explain why an optimum IT1 duration (60 sec. in the present experiment) produces a deeper response decrement during training than shorter or longer ones. However, if the dual-process theory can appropriately account for the observed results during habituation training, the data dealing with retention performance after the rest period can hardly be explained within this theoretical framework. The best measure of habituation retention is given by the responsiveness
of the animals to the
first stimulation of the retention session, since sensitixing effects are not yet triggered by stimulus repetition. The retention of habituation appears to be better when the IT1duration was longer during previous training. Three hypotheses could be suggested to account for this result: I I a residual sensitization. the level of which is higher the shorter the ITI, still remains after the rest period. 2) short ITIs impair the development of habituation process during training. 3) even if habituation took place during training, short ITIs prevent its long-term retention. The first two hypothesis imply that a correlation could be found between training performance, evaluated by the final response level. and retention score, estimated by response likelihood after the rest period. Such a correlation cannot be found, since it is impossible to predict if an animal will respond to the visual stimulation following the rest period from its response level at the end of training (Fig. 2). Moreover, ani-
55
mals which respond and those which do not respond to that stimulation display exactly the same pattern of response variation during training
session (Fig. 31. This lack
of correlation between acquisition and retention performance has been often reported (Then. 1981;Thon and Pat&, 19841and suggests an independence of the processes underlying
short-term and long-term retention of habituation. This independence is
also supported by comparing the scores of ITI:
and ITI
groups: while these
groups reach the same response level at the end of training, response likelihood after the rest period raises up in the ITI: suggesting some reminiscence-like
animals and falls down in the ITI
group,
process in the latter. These arguments support
the first hypothesis which postulates a differential effect of ITI duration on shortterm and long-term retention of habituation.
Davis (1970aI suggests that short ITIs
prevent the consolidation of habituation in long-term memory: habituation, triggered by a given stimulation would be disrupted if the next stimulation is applied after a short interval. Other experimental data show that response decrement observed with high frequency
of stimulation results from refractory-like
processes (Wilson and
Groves, 1973) which could impair the development of long-term habituation. According to Wagner (1976,19791, a lack of responding to a stimulus results from a priming of the stimulus representation
into short-term memory storage. This pri-
ming can be induced by a recent occurence of that stimulus (self-generated priming) or by the presentation of environmental
cues previously associated with the stimulus
during initial acquisition session (retrieved-generated
priming). Self-generated pri-
ming is involved in short-term retention of habituation and plays an important part in response decrement during training, while retrieved-generated
priming is a de-
terminant of long-term retention of habituation (Whitlow, 19751.The development of the association between contextual cues and the to-be-habituated stimulus would be improved by long intervals between successive trials. Even if the presents results can hardly be accounted for by this theoretical model, since the animals were only presented with the releasing stimulus with a few -if any- environmental
cues, they sug-
gest that behavioural habituation in the blowfly involves several independent cesses during acquisition and retention, and that memory-centered to be restricted to vertebrates, but could be generalized to invertebrate
pro-
models have not habituation.
Another point which deserves attention in the present results, is the tendency for a transient rise in response level during the second session by animals which are not responsive at the end of the rest period. Such a phenomenon
has been previously
described in the blowfly (Than and Pauzie, 19841,and seems not to be an artifact. It appears to be independent of IT1 duration (Fig. 3). It can be considered as the result of the development of sensitization when the stimulation is repeated again. We can presume that this transient sensitization cannot be revealed in responding animals because of ceiling effects. No explanation of this phenomenon
can be proposed, but it
56
stresses the importance of the analysis of individual pattern of response variation in habituation studies, which reveales interesting
features generally hiden when the
data are averaged within experimental groups. REFERENCES Askew, HR, 1970. Effects of stimulus intensity and intertrial interval on habituation of the head-shake response in the rat. J. Camp. Physiol. PsychoI.. 72: 492497, Barras, R., 1961. A quantitative study of the behaviour of the male Mormoniella u (Hym. pteromalidae) towards two constant stimuli situations, Be2 viour, 18:288312. Brzoska, J., 1983. Amplitude, latency and habituation of the electrodermal response to acoustic stimuli in the frog. Behav. Proc., 8: 229-242. Carew. T.J. and Kandel, E.R.. 1973. Acquisition and retention of long-term habituation in Aplysia: correlation of behavioral and cellular processes. Science. 182: 1158-1160. Carew, T.J., Pinsker, H.M.and Kandel, E.R.1972.Long-term habituation of a defensive withdrawal1 reflex in At,&&. Science, 175.451-454. Cook, A., 1971. Habituation in the freshwater snail tm MI. Anim. Behav., 19:463-474. Crawford, F.T., 1979.The effect of distribution of trials upon the habituation of tonic immobility in the tarentula mm. Bull. Psychon. Sot., 14: 135-137. Davis, M., 1970a. Effect of interstimulus length and variability on startle response habituation in the rat. J. Camp. Physiol. Psychol., 72: 177-192. Davis, M, 1970b. Interstimulus interval and startle response habituation with a ‘control’ for total time during training. Psychon. Sci.. 20: 39-41. Davis, M., 1972.Differential retention of sensitization and habituation of the startle response in the rat. J. Camp. Physiol. Psychol., 78: 260-267. Davis M. and Henninger, G.R..1972.Comparison of response plasticity between the eyeblink and vertex potential in humans. Electoencephal. Clin. Neurophysiol., 33:283-293. File, SE., 1973. Interstimulus interval and the rate of behavioural habituation. Quart. J. Exp. Psychol., 25: 360-367. Gatchel, R.J., 1975.Effects of interstimulus interval length on short- and long-term habituation of autonomic components of the orienting response. Physiol. Psychol., 3: 133-136. Gatchel, R.J. and Lang, P.J., 1974. Effects of interstimulus interval length and variability on habituation of autonomic components of the orienting response. J. Exp. Psychol., 103: 802-804. GIantman. D.L.and Schmidt, EC., 1979.Habituation of the nictitating membrane reflex in the intact frog. Physiol. Behav.. 22: 1141-1148. Groves, P.M., Lee, D.,and Thompson, R.F., 1969.Effects of stimulus frequency and intensity on habituation and sensitization in acute spinal cat. Physiol. Behav.. 4: 383-388. Groves, P.M. and Thompson, R.F.. 1970. Habituation: a dual-process theory Psychol. Rev., 77: 419450. Horn, G., 1%7. Neuronal mechanisms of habituation. Nature, 215: 707-711. Laming, P.R. and Ennis, P., 1982. Habituation of fright and arousal in the teleosts Csrassiusmand&&&&ifhs J. Camp. Physiol. Psychol., 96: 460-466. Patterson, T.L. and Petrinovich. L., 1979.Field studies of habituation. II) Effects of massed stimulus presentation. J. Camp. Physiol. Psychol., 93: 351-359. Peeke, H.V.S.and Peeke, SC, 1970. Habituation of conspecific aggressive responses in the Siamese fighting fish &&&solende&. Behaviour. 36: 232-245. Petrinovich, L., 1984.A two-factor dual-process theory of habituation and sensitiza-
57
tion. In H.V.S.Peeke and L. Petrinovich, (Edititorsl, Habituation, Sensitization and Behavior, Academic Press, New York, pp 17-55. Pieron. II., 1909. L’adaptation aux obscurations repetees comme phenomene de memoire chez les animaux inferieurs. Arch.Psychol., 9: 39-50. Pieron. II.. 1911.Sur la determination de la periode d’etablissement dans les acquisitions mnemoniaues. C.R. Acad. Sci. Paris, 152: 1410-1411. Sanes, J.N. and Ison,-J.R., 1983. Habituation and-sensitixation of components of the human eyeblink reflex. Behav. Neurosci., 97: 833-836. Sokolov. E.N., 1%3. Perception and the conditional reflex. Pergamon Press, New York. Stein, L., 1966. Habituation and stimulus novelty: a model based on classical conditionning. Psychol. Rev., 73: 352356. Terry, W.S., 1979. Habituation and dishabituation of rat’s exploration of a novel environment. Anim. Learn. Behav.. 7: 525-536. Thompson, R.F.. Groves, P.M., Teyler, T.J. and Roemer, R.A., 1973. A dual-process theory of habituation: theory and behavior. In H.V.S.Peeke and M.J. Hen (Editors). Habituation. Vol.1. Behavioral Studies. Academic press, New York, pp 239-271. Thompson R.F. and Spencer, W.A., 1%6. Habituation: a model phenomenon for the study of neuronal substrates of behavior. Psychol. Rev., 173: 16-43. Thon, B.. 1980.Habituation of cardiac and motor responses to a moving visual stimulus in the blowfly. J. Comp. Physiol. Psychol., 94: 886-893. Thon, B.. 1981. Etude quantitative de la retention de l’habituation de la reponse optocardiaque chet Calliohora vomitoria. Physiol. Behav., 26: 423-432. Thon, B.. 1982. Influence of the cardiac phase on the latency of a motor response to avisual stimulus in the blowfly. J. Insect Physiol., 28: 411-416. Thon, B. and Pauzie, A., 1984. Differential sensitization, retention and generalization of habituation in two response systems in the blowfly (Calliohoravomitod). J. Comp.Psychol., 98: 119-130. Thon, B. and Queinnec, Y., 1976.Habituation de la reponse optocardiaque chex calli_ W~Q&Q&. Biol. Behav., 2: 101-113. Wagner, A.R.. 1976. Priming in STM: an information processing mechanism for self-generated and retrieved-generated depression in performance. In: T.J. Tighe and N. Leaton (Editors). Habituation: perspectives from child development, animal behavior and neurophysiology. Lawrence Erlbaum Ass.. Hillsdale, pp. 95-128. Wagner, A.R., 1979. Habituation and memoryIn: A. Dickinson and R.A. Boakes (Editors). Mechanisms of learning and motivation. A memorial volume for J. Konorski. Lawrence Erlbaum Ass., Hillsdale, pp. 53-82. Whitlow, J.W.. 1975. Short-term memory in habituation and dishabituation. J. Exp. Psychol. Anim. Behav. Proc., 104: 189-206. Whitlow, J.W. and Wagner, A.R., 1984. Memory and habituation. In H.V.S.Peeks and L. Petrinovich, (Editors), Habituation, Sensitixation and Behavior. Academic Press, New York, pp. 103-153. Wilson, C.J. and Groves, P.M., 1973. Refractory period and habituation of acoustic startle response in the rat. J. Comp. Physiol. Psychol., 83: 492-498. Wirier. J., 1971. Statistical principles in experimental design. McGraw Hill, New York. Wood,DC., 1970.Parametric study of the response decrement produced by mechanical stimuli in the protozoan Stentor coeruleus J. Neurobiol., 1: 345-360.