- Email: [email protected]
Regulatory mechanisms underlying novelty-induced grooming in the laboratory rat
Behavioural Processes 67 (2004) 287–293
Regulatory mechanisms underlying novelty-induced grooming in the laboratory rat Joanna Komorowska∗ , Sergio M. Pellis Department of Psychology and Neuroscience, Canadian Centre for Behavioural Neuroscience, University of Lethbridge, 4401 University Drive, Lethbridge, Alta., Canada T1K 3M4 Received 26 April 2004; accepted 6 May 2004
Abstract Bouts of pelage cleaning can be readily evoked in laboratory rodents under conditions of exposure to novelty. Such noveltyinduced grooming is described as stereotyped and rostro-caudal in its progression. The patterned structure of novelty-induced grooming makes it particularly attractive for research on the organizational and motivational underpinnings of co-ordinated behaviour. Micro-characteristics of stereotyped novelty-induced grooming bouts were studied in 27 female Wistar rats that were exposed to a novel arena with shelters for a period of 15 min. The order of grooming acts within the initial bouts was rostro-caudal, but subsequent bouts became progressively disorganized in their sequencing. The observed pattern of progressive bout disorganization may be attributed to the gradual dearousal from stress. Differences between consecutive bouts in their micro-characteristics suggest that at least some grooming actions emitted within the context of those bouts operate as relatively independent units of behaviour. Those unitary component actions appear to be integrated into protracted pelage cleaning sequences by a separate mechanism. Similar endogenous mechanisms have been proposed for other co-ordinated motor actions, which suggests that the organizational principles identified in the context of novelty-induced grooming may represent general principles that govern co-ordinated behaviour. © 2004 Elsevier B.V. All rights reserved. Keywords: Grooming; Novelty; Rostro-caudal progression; Sequential stereotypy; Grooming bout organization; Patterned behaviour
1. Introduction Co-ordinated behaviour is believed to be a product of an interplay between mechanisms that operate on many levels of neural organization (Gallistel, 1980). This means that structural and temporal characteristics of behaviour may provide clues as to the organizational and motivational principles that guide such behaviour (e.g. Fentress, 1972). Behaviours that occur naturally (i.e. do not require extensive training), that ∗ Corresponding author. Tel.: +1 403 3943951; fax: +1 403 3292555. E-mail address: [email protected] (J. Komorowska).
have a diverse, patterned structure and that are reliably evoked under predictable conditions are the ones particularly suitable for this purpose (e.g. Grillner, 1975). Grooming in rodents meets the above criteria. It incorporates a collection of diverse component actions that are strung together into temporally discrete and structurally diverse sequences or bouts (e.g. Berridge, 1989a,b; Fentress, 1972). Aside from the obvious circumstance of pelage maintenance and care, grooming is predictably triggered by environmental disturbance, for example under conditions of novelty exposure (Spruijt et al., 1992; van Erp et al., 1994). Novelty-induced grooming bouts are reported as stereotyped and rostro-caudal in their sequential
0376-6357/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.beproc.2004.05.001
288
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
organization: they are initiated with face directed actions and progress towards the posterior parts of the body (Fentress, 1972; Richmond and Sachs, 1980; Sachs, 1988; Spruijt and Gispen, 1983). One particular cluster of grooming actions that is characterized by an extraordinarily high degree of sequential stereotypy is the syntactic chain (e.g. Berridge, 1989b; Berridge and Fentress, 1986; Berridge and Fentress, 1987). It is emitted within a bout of grooming immediately after an initial phase of preliminary rostral strokes and paw licks (Berridge and Fentress, 1987) and incorporates a collection of diverse actions, including an easily definable cascade of rapid, bilaterally symmetrical wipes over the mystacial vibrissae. The syntactic chain is terminated with a head tucking movement that marks a transition between the rostral and the caudal stage of the grooming bout. Grooming then continues ventro-laterally and gradually shifts towards the most posterior parts of the body (Richmond and Sachs, 1980; Sachs, 1988). The sequential stereotypy of the syntactic chain is described as exceptionally rigid and unaffected by peripheral stimulation (Berridge, 1989a; Berridge and Fentress, 1986). The amount of novelty-induced grooming has been reported to change over the time-period following the stimulus presentation (van Erp et al., 1994). This effect is due, at least in part, to the gradual increase in the duration of consecutive bouts during the course of novelty-exposure (Choleris et al., 2001; Fentress, 1972; Horvath et al., 1971; Hughes, 1991; Komorowska and Pisula, 2003; Pleskacheva, 1996). Considering the diverse nature of grooming sequences, it is possible that the microstructure of the stereotyped grooming bouts may also undergo predictable changes under those circumstances. The fact that novelty-induced grooming bouts are rostro-caudal in their progression and that the transition from the rostral to the caudal phase is easily defined by the presence of the syntactic chain, makes it possible to scrutinize the content of the bouts for any subtle between-bout differences. The nature of those differences may shed light on the endogenous mechanisms that are involved in the production of the pelage-cleaning behaviour. For example, the increase in the overall duration of consecutive bouts may be accompanied by a simultaneous increase in the duration of certain component grooming actions. Con-
versely, the duration of component actions may remain the same for all bouts. Each one of the two possibilities will lead to different conclusions regarding the mechanisms that are responsible for the expression of the grooming behaviour. We attempted to gain insight into the nature of those mechanisms, by analysing the micro-characteristics of consecutive grooming bouts emitted in a novel environment.
2. Method Subjects were 27 na¨ıve Wistar female rats obtained from a breeder (Szostak; Warsaw, Poland) at 40–50 days of age. Females were used for practical reasons of availability, because there is no convincing evidence for reliable sex differences in stress-related grooming (e.g. Moore, 1986; Pleskacheva, 1996). The subjects were housed in groups of 3 and 4 in wire mesh cages (44 cm × 28 cm × 28 cm) at the Warsaw University Psychology Department animal colony, under standard laboratory conditions (12-light:12-dark cycle) on commercial rat chow diet and water ad libitum. The animals were tested at circa 83 days of age (Komorowska and Pisula, 2003). 2.1. Apparatus The testing apparatus consisted of an illuminated rectangular open arena (120 cm × 150 cm) and two shelters made of transparent glass, positioned in the centre of the arena. The surface of the arena was covered with grey linoleum, light was of moderate intensity (30 lx as measured at the surface of the arena). 2.2. Procedure On the day of testing, each animal was transported into the experimental room in an opaque plastic container and left undisturbed for a 10 min period. Afterwards, the rat was picked up by the experimenter (person known to subjects) and positioned in the apparatus for a period of 15 min. (It was established during preliminary tests that 15 min were sufficient for subjects to greatly reduce their activity in the apparatus.) For standardization purposes each rat was placed directly into one of the shelters. Behaviour of the subject animals was videotaped with a VHS
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
289
camera (Panasonic) and scored from the videotaped recordings. The surface of the arena and the interior of the shelters were thoroughly washed with soapy water and dried between subjects. 2.3. Description of behavioural categories Of all pelage oriented activities, the stereotyped grooming bouts were chosen for analyses, that is, those bouts that incorporated the syntactic chain (see Section 1 for details). Pauses in grooming longer than 8 s in duration were treated as a cut-off point between separate bouts. All analyses were performed on the first three grooming bouts, since further bouts were too infrequent to warrant their statistical comparison. Two general categories of within-bout activities were identified. These were: (A) Rostral grooming (RG): All actions performed by the forepaws and directed at the snout, face and head areas, including the ears. Within the category of rostral grooming a subcategory of preliminary rostral grooming (PRG) was identified. This subcategory incorporated forepaw licking and the wiping of the snout area. (B) Caudal grooming (CG): Pelage cleaning of the trunk and the extremities. This category incorporated the actions of licking, scratching and/or nibbling the fur of the belly, the flanks and the back, the anogenital area and the hind paws as well as the furry parts of the forelimbs (arm and/or shoulder). Within the general category of caudal grooming a subcategory of lateral caudal grooming (LCG) was identified that involved the licking, rubbing and/or nibbling of the flanks and the dorsal parts of the trunk. Each bout was divided into two segments (S1 and S2), with the syntactic chain marking the end of S1 and the beginning of S2 (Fig. 1). Three parameters pertaining to the S1 segment of each bout were recorded. These were: (1) total duration (measured as the number of 3 s intervals), (2) duration of uninterrupted preliminary rostral grooming (measured as the number of consecutive 3 s intervals), (3) presence of caudal grooming (probability of occurrence). Similarly, two bout parameters pertaining to the S2 segment were recorded. These were: (1) total duration (measured as the number of 3 s intervals),
Fig. 1. A schematic representation of a stereotyped grooming bout. The initial section (terminated by the syntactic chain) represents the S1 segment of the bout, the end section represents the S2 segment of the bout.
(2) duration of uninterrupted lateral caudal grooming (measured as the number of consecutive 3 s intervals). If more than one episode of lateral caudal grooming was present within a given bout, the longest episode was used for comparisons. 2.4. Statistical analyses Sample variances lacked homogeneity, therefore, the non-parametric Friedman analysis of variance was used for statistical comparisons. Subsequent pairwise comparisons were performed using the Wilcoxon matched-pair signed-ranks test (using a P < 0.05 cut-off).
3. Results There were significant differences in the duration of consecutive bouts for both S1 (χr2 (2) = 14.63, P < 0.001), and S2 (χr2 (2) = 13.63, P < 0.001, Fig. 2). Pairwise comparisons for both S1 and S2 revealed differences between bouts one and two and between bouts one and three, but not between bouts two and three. Between-bout differences were present in the duration of uninterrupted preliminary rostral grooming (χr2 (2) = 13.3, P < 0.001, Fig. 3A). PRG was significantly shorter for bout one than for bout two, it was also shorter for bout one than for bout three. No significant difference was present between-bout two and bout three. The duration of lateral caudal grooming changed in the similar manner (χr2 (2) = 20.83, P < 0.001, Fig. 3B). It was the shortest in the context of the first bout and increased in its duration for subsequent bouts. All pairwise comparisons revealed significant differences between bouts.
290
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
Fig. 2. Duration of segment S1 (left) and segment S2 (right) of consecutive grooming bouts. Data represent mean ± S.E.M.
The probability of caudal grooming within the S1 segment of a bout differed between consecutive bouts (χr2 (2) = 20.83, P < 0.05, Fig. 3C). Caudally directed intrusions into S1 were significantly less likely within bout one than within bout two and significantly less likely within bout one than within bout three. No significant difference was revealed between bouts two and three.
4. Discussion The duration of the S1 segment differed between consecutive bouts. The S1 of the initial bout was the shortest, while the S1 of the third bout was the longest. The increase in the duration of segment 1 appeared to arise as a result of two co-occurring events: the appearance of caudal grooming in the S1 segment of
Fig. 3. Micro-characteristics of the consecutive grooming bouts: duration of preliminary rostral grooming (A), duration of uninterrupted caudal grooming (B), presence of caudal grooming within the S1 segment of a bout (C). All data represent mean ± S.E.M.
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
the later occurring bouts and a prolonged period of preliminary rostral grooming within the framework of those bouts. The observed differences in bout characteristics shed light on the regulatory mechanisms that are involved in the control of grooming sequences. Bouts of novelty-induced grooming are believed to be stereotyped, with component grooming acts progressing in an orderly snout-face-head-trunk sequence (e.g. Richmond and Sachs, 1980; Sachs, 1988; Spruijt and Gispen, 1983). Our data do not fully support this general assumption. It is true that the first grooming bouts followed the rostro-caudal progression: such bouts started rostrally and proceeded smoothly past the rostro-caudal threshold of the syntactic chain. However, in the context of later bouts, the presence of caudal grooming prior to the occurrence of the syntactic chain indicates a disruption of the rostro-caudal pattern. This suggests that the rostro-caudal order of grooming actions may be less typical than is generally acknowledged. The presence of ‘disorganized’ bouts within the trial can not be interpreted as an inconsequential or incidental deviation from the rostro-caudal sequence. On the contrary, the fact that the disorganized bouts tended to follow the rostro-caudally organized bouts indicates an orderly transition between both types of bouts. It appears that the differences between both bout types in their sequential patterning may reflect a shift in the internal state of the grooming animal. Various lines of evidence point to such shift as being a reflection of a decrease in the level of stress. Neurophysiological studies demonstrate that the exposure to novelty triggers the stress response (e.g. Armario et al., 1986; De Boer et al., 1990; Pfister, 1979). They also indicate that the intensity of this response depends on the intensity of the stressor (Armario et al., 1986; De Boer et al., 1990) and that the habituation to novelty reduces the neurophysiological stress (De Boer et al., 1990; Kant et al., 1984; van Erp et al., 1994; Roth and Katz, 1979). On this basis, the initial conditions of the present study can be assessed as moderately stressful with subsequent attenuation of the stress response taking place during the course of the trial. Many behavioural variables are known for their low reliability as markers of stress and stress-related states (e.g. Rodgers, 1997). However, one type of risk assessment behaviour: the stretched-attend posture (SAP) stands out as a robust correlate of physiological stress.
291
For example, high frequency of SAP is associated with high corticosterone levels and elevated blood pressure (Rodgers, 1997; Dielenberg and McGregor, 2001; Rodgers et al., 1999) and it can be reduced by repeated exposures to novelty (Kaesermann, 1986; Choleris et al., 2001). Interestingly, high SAP scores have also been shown to immediately precede initial but not later occurring bouts of grooming (Komorowska and Pisula, 2003).This last result demonstrates that the first bouts of grooming performed in a novel arena are associated with higher physiological stress than the subsequent bouts. Overall, the available evidence suggests that the difference in the organizational pattern of consecutive bouts (their progressive disorganization) may be due, at least in part, to a gradual dearousal from stress. The attribution of bout disorganization to a decrease in the level of stress provides a clue as to why the abundance of non-rostro-caudal bouts has not been adequately acknowledged. Typical experimental conditions involve manipulations on the subject animals’ direct environment immediately prior to the recording of their behavioural activity. Evidence demonstrates that even those procedures that are believed to be relatively benign, for example the opening of the cages or the shuffling around of the subjects, cause noticeable elevation of the neurophysiological measures of stress (Rollin, 1998). It may be that under typical experimental conditions, the triggering of stress promotes the rostro-caudal organization of the grooming bouts. If low stress conditions are indeed required to promote bout disorganization, no disorganized bouts will be present under the usual experimental settings. In agreement with this interpretation, and contrary to the typical findings, Bolles (1960) observed that bouts of grooming emitted by rats in the undisturbed setting of home cages did not progress rostro-caudally. The second finding of the present study relates to the changes in the duration of the two constituent segments of consecutive bouts. Both segments were the shortest for the first bout and gradually increased in length for the following bouts. The fact that the increase in the duration of one segment was coupled with a simultaneous increase in the duration of the other segment means that entire bouts were increasing in their length over the course of the trial. Gradual prolongation of bouts has also been reported for other novel settings (in mice: Choleris et al., 2001 and
292
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
Fentress, 1972; in rats: Horvath et al., 1971; Hughes, 1991 and Pleskacheva, 1996) and may be related to the decrease in the level of stress over the course of novelty exposure (Choleris et al., 2001; Komorowska and Pisula, 2003; Pleskacheva, 1996). Interestingly, within the context of the present study, not only bouts themselves but also two component activities within those bouts changed in their overall duration. Both components: the preliminary rostral grooming and the lateral caudal grooming were emitted in the longest uninterrupted strings of activity within the final bouts. The pattern of changes observed for the two component activities suggests the involvement of two mechanisms in the shaping of these activities. A mechanism of activation was responsible for the initiation of a particular action, and a mechanism of continuation allowed the given action to be emitted for a period of time before it was replaced with another component action. Clearly, since both PRG and LCG were present in all analysed bouts, the mechanism of activation was operating unimpeded within the framework of all bouts. However, a relatively short duration of the two component actions within the first bouts suggests that the process of continuation was disturbed in the context of those bouts, but not in subsequent bouts. This dichotomy suggests that the two mechanisms (activation and continuation) should be viewed as operating relatively independently of one another. The results also point to another organizational property that operates on grooming behaviour. Since both PRG and LCG could be lengthened or shortened within the broader framework of individual grooming bouts, they should be treated as relatively independent units of those bouts. However, and despite their relative independence, the two activities were still part and parcel of the structurally diverse and fluidly executed sequences of motor actions. This suggests another mechanism—this mechanism is responsible for the integration of the grooming activities into uninterrupted and temporally discrete pelage cleaning sequences. It appears that grooming bouts constitute an end product of interplay between processes of activation and integration of constituent grooming activities. Research on the early development of face grooming in laboratory rodents lends support to the above interpretation. It points to various face directed strokes as loosely structured during the initial stages of devel-
opment. Originally, such actions tend to be performed separately and independently of each other (Fentress, 1972; Golani and Fentress, 1985). Only later in development, does this independence disappear and the unitary motor patterns become integrated into longer sequences of actions (Golani and Fentress, 1985). The fact that the mechanisms of activation and integration of component actions have been identified both for mature expression of novelty-triggered grooming and for the development of pelage care highlights the importance of both mechanisms in the organization of the grooming behaviour. An initial independence of motor actions and the subsequent integration of those component-actions, have also been observed during the development of several other naturally occurring behaviours (e.g. righting in rodents and marsupials: Pellis et al., 1992 and Pellis et al., 1991; locomotor co-ordination in rodents: Bekoff and Trainer, 1979).The interesting parallel between our findings regarding novelty-induced grooming and the findings regarding other patterned behaviours suggests that the interplay between the activation and the integration of the unitary component actions is not limited to pelage cleaning. This interplay may represent a fundamental principle that guides the production of patterned behaviour. In conclusion, the present results reveal that a number of regulatory processes are involved in the expression of novelty-induced grooming. Grooming starts rostro-caudally but becomes progressively disorganized as the trial continues—a pattern that may be due to a shift in the motivational state of the grooming animal, possibly a gradual dearousal from stress. The differences in the micro-characteristics of consecutive grooming bouts suggest the involvement of specific endogenous mechanisms in the shaping of those bouts. It appears that at least some component activities should be considered as units whose activation and continuation are controlled independently within the framework of individual grooming bouts. A mechanism of integration appears to be necessary to bind those component actions into cohesive, uninterrupted and protracted sequences of pelage cleaning. The organizational principles identified in the context of novelty-induced grooming may, in fact, be representative of fundamental organizational principles that govern co-ordinated behaviour in general.
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
References Armario, A., Montero, J.L., Balasch, J., 1986. Sensitivity of corticosterone and some metabolic variables to graded levels of low intensity stresses in adult male rats. Physiol. Behav. 37, 559–561. Bekoff, A., Trainer, W., 1979. The development of interlimb co-ordination during swimming in postnatal rats. J. Exp. Biol. 83, 1–11. Berridge, K.C., 1989a. Progressive degradation of serial grooming chains by descending decerebration. Behav. Brain Res. 33, 241– 253. Berridge, K.C., 1989b. Substantia nigra 6-OHDA lesions mimic stratopallidal disruption of syntactic grooming chains: a neural systems analysis of sequence control. Psychobiology 17, 377– 385. Berridge, K.C., Fentress, J.C., 1986. Contextual control of trigeminal sensorimotor function. J. Neurosci. 9, 325–362. Berridge, K.C., Fentress, J.C., 1987. Disruption of natural grooming chains after stratopallidal lesions. Psychobiology 15, 336–342. Bolles, C., 1960. Grooming behaviour in the rat. J. Comp. Physiol. Psychol. 53, 306–310. Choleris, E., Thomas, A.W., Kavaliers, M., Prato, F.S., 2001. A detailed ethological analysis of the mouse open field test: effects of diazepam, chloriazepoxide and an extremely low frequency pulsed magnetic field. Neurosci. Biobehav. Rev. 25, 235– 260. De Boer, S.F., Koopmans, S.J., Slangen, J.L., van der Gugten, J., 1990. Plasma catecholamine, cortocosterone and glucose responses to repeated stress in rats: effect of interstressor interval length. Physiol. Behav. 47, 1117–1124. Dielenberg, R.A., McGregor, I.S., 2001. Defensive behaviour in rats towards predatory odors: a review. Neurosci. Biobehav. Rev. 25, 597–609. Fentress, J.C., 1972. Development and patterning of movement sequences in inbred mice. In: Kiger, J. (Ed.), The Biology of Behaviour. Oregon State University Press, Oregon, pp. 83– 132. Gallistel, C.R., 1980. The Organization of Action: A New Synthesis. Erlbaum Associates, Hillsdale, 432 pp. Golani, I., Fentress, J.C., 1985. Early ontogeny of face grooming in mice. Dev. Psychobiol. 18, 529–544. Grillner, S., 1975. Locomotion in vertebrates: central mechanisms and reflex interactions. Physiol. Rev. 55, 274–385. Horvath, T., Kirby, H.W., Smith, A.A., 1971. Rat’s heart rate and grooming activity in the open field. J. Comp. Physiol. Psychol. 73, 449–453. Hughes, R.N., 1991. The role of self- and other-animal-produced odors, in rats’ preferences for novelty, in an exploration box. Psychobiology 19, 168–174.
293
Kaesermann, H.P., 1986. Stretched attend posture, a non-social form of ambivalence, is sensitive to a conflict-reducing drug action. Psychopharmacology 89, 31–37. Kant, G., Eggleston, T., Landman-Roberts, L., Kenion, C., Driver, G., Meyerhoff, J., 1984. Habituation to repeated stress is stressor specific. Pharmacol. Biochem. Behav. 22, 631–634. Komorowska, J., Pisula, W., 2003. Does changing levels of stress affect the characteristics of grooming behaviour in rats? Int. J. Comp. Psychol. 16, 237–246. Moore, C.L., 1986. Sex differences in self-grooming of rats: effects of gonadal hormones and context. Dev. Psychobiol. 17, 243– 253. Pellis, S.M., Pellis, V.C., Nelson, J.E., 1992. The development of righting reflexes in the pouch young of the marsupial Dasyurus hallucatus. Dev. Psychobiol. 25, 105–125. Pellis, V.C., Pellis, S.M., Teitelbaum, P., 1991. A descriptive analysis of the postnatal development of contact-righting in rats. Dev. Psychobiol. 24, 237–263. Pfister, P., 1979. The glucocorticosterone response to novelty as a psychological stressor. Physiol. Behav. 23, 649–652. Pleskacheva, M.G., 1996. Temporal characteristics of grooming in an open field in two strains of rats. Int. J. Comp. Psychol. 9, 105–117. Rodgers, R., 1997. Animal models of ‘anxiety’: where next? Behav. Pharmacol. 8, 477–496. Rodgers, R.J., Haller, J., Holmes, A., Halasz, J., Walton, T.J., Brain, P.F., 1999. Corticosteron response to the plus-maze: high correlation with risk assessment in rats and mice. Physiol. Behav. 68, 47–53. Richmond, G., Sachs, B.D., 1980. Grooming in Norway rats: the development and adult expression of a complex motor pattern. Behaviour 75, 82–96. Rollin, B.E., 1998. The Unheeded Cry: Animal Consciousness, Animal Pain, and Science. Iowa State University Press, Ames, 330 pp. Roth, K., Katz, R., 1979. Stress, behavioural arousal, and open field activity – a reexamination of emotionality in the Rat. Neurosci. Biobehav. Rev. 3, 247–263. Sachs, B.D., 1988. The development of grooming and its expression in adult animals. Ann. N.Y. Acad. Sci. 525, 1–17. Spruijt, B., Gispen, W., 1983. Behavioural sequences as an easy quantifiable parameter in experimental studies. Physiol. Behav. 32, 707–710. Spruijt, B., van Hooff, J., Gispen, W., 1992. The ethology and neurobiology of grooming behaviour. Physiol. Rev. 72, 825– 852. van Erp, A., Kruk, M., Meelis, W., Willekens-Bramer, D., 1994. Effect of environmental stressors on time course, variability and form of self-grooming in the rat: handling, social contact, defeat, novelty, restraint and fur moistening. Behav. Brain Res. 65, 47–55.
Regulatory mechanisms underlying novelty-induced grooming in the laboratory rat Joanna Komorowska∗ , Sergio M. Pellis Department of Psychology and Neuroscience, Canadian Centre for Behavioural Neuroscience, University of Lethbridge, 4401 University Drive, Lethbridge, Alta., Canada T1K 3M4 Received 26 April 2004; accepted 6 May 2004
Abstract Bouts of pelage cleaning can be readily evoked in laboratory rodents under conditions of exposure to novelty. Such noveltyinduced grooming is described as stereotyped and rostro-caudal in its progression. The patterned structure of novelty-induced grooming makes it particularly attractive for research on the organizational and motivational underpinnings of co-ordinated behaviour. Micro-characteristics of stereotyped novelty-induced grooming bouts were studied in 27 female Wistar rats that were exposed to a novel arena with shelters for a period of 15 min. The order of grooming acts within the initial bouts was rostro-caudal, but subsequent bouts became progressively disorganized in their sequencing. The observed pattern of progressive bout disorganization may be attributed to the gradual dearousal from stress. Differences between consecutive bouts in their micro-characteristics suggest that at least some grooming actions emitted within the context of those bouts operate as relatively independent units of behaviour. Those unitary component actions appear to be integrated into protracted pelage cleaning sequences by a separate mechanism. Similar endogenous mechanisms have been proposed for other co-ordinated motor actions, which suggests that the organizational principles identified in the context of novelty-induced grooming may represent general principles that govern co-ordinated behaviour. © 2004 Elsevier B.V. All rights reserved. Keywords: Grooming; Novelty; Rostro-caudal progression; Sequential stereotypy; Grooming bout organization; Patterned behaviour
1. Introduction Co-ordinated behaviour is believed to be a product of an interplay between mechanisms that operate on many levels of neural organization (Gallistel, 1980). This means that structural and temporal characteristics of behaviour may provide clues as to the organizational and motivational principles that guide such behaviour (e.g. Fentress, 1972). Behaviours that occur naturally (i.e. do not require extensive training), that ∗ Corresponding author. Tel.: +1 403 3943951; fax: +1 403 3292555. E-mail address: [email protected] (J. Komorowska).
have a diverse, patterned structure and that are reliably evoked under predictable conditions are the ones particularly suitable for this purpose (e.g. Grillner, 1975). Grooming in rodents meets the above criteria. It incorporates a collection of diverse component actions that are strung together into temporally discrete and structurally diverse sequences or bouts (e.g. Berridge, 1989a,b; Fentress, 1972). Aside from the obvious circumstance of pelage maintenance and care, grooming is predictably triggered by environmental disturbance, for example under conditions of novelty exposure (Spruijt et al., 1992; van Erp et al., 1994). Novelty-induced grooming bouts are reported as stereotyped and rostro-caudal in their sequential
0376-6357/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.beproc.2004.05.001
288
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
organization: they are initiated with face directed actions and progress towards the posterior parts of the body (Fentress, 1972; Richmond and Sachs, 1980; Sachs, 1988; Spruijt and Gispen, 1983). One particular cluster of grooming actions that is characterized by an extraordinarily high degree of sequential stereotypy is the syntactic chain (e.g. Berridge, 1989b; Berridge and Fentress, 1986; Berridge and Fentress, 1987). It is emitted within a bout of grooming immediately after an initial phase of preliminary rostral strokes and paw licks (Berridge and Fentress, 1987) and incorporates a collection of diverse actions, including an easily definable cascade of rapid, bilaterally symmetrical wipes over the mystacial vibrissae. The syntactic chain is terminated with a head tucking movement that marks a transition between the rostral and the caudal stage of the grooming bout. Grooming then continues ventro-laterally and gradually shifts towards the most posterior parts of the body (Richmond and Sachs, 1980; Sachs, 1988). The sequential stereotypy of the syntactic chain is described as exceptionally rigid and unaffected by peripheral stimulation (Berridge, 1989a; Berridge and Fentress, 1986). The amount of novelty-induced grooming has been reported to change over the time-period following the stimulus presentation (van Erp et al., 1994). This effect is due, at least in part, to the gradual increase in the duration of consecutive bouts during the course of novelty-exposure (Choleris et al., 2001; Fentress, 1972; Horvath et al., 1971; Hughes, 1991; Komorowska and Pisula, 2003; Pleskacheva, 1996). Considering the diverse nature of grooming sequences, it is possible that the microstructure of the stereotyped grooming bouts may also undergo predictable changes under those circumstances. The fact that novelty-induced grooming bouts are rostro-caudal in their progression and that the transition from the rostral to the caudal phase is easily defined by the presence of the syntactic chain, makes it possible to scrutinize the content of the bouts for any subtle between-bout differences. The nature of those differences may shed light on the endogenous mechanisms that are involved in the production of the pelage-cleaning behaviour. For example, the increase in the overall duration of consecutive bouts may be accompanied by a simultaneous increase in the duration of certain component grooming actions. Con-
versely, the duration of component actions may remain the same for all bouts. Each one of the two possibilities will lead to different conclusions regarding the mechanisms that are responsible for the expression of the grooming behaviour. We attempted to gain insight into the nature of those mechanisms, by analysing the micro-characteristics of consecutive grooming bouts emitted in a novel environment.
2. Method Subjects were 27 na¨ıve Wistar female rats obtained from a breeder (Szostak; Warsaw, Poland) at 40–50 days of age. Females were used for practical reasons of availability, because there is no convincing evidence for reliable sex differences in stress-related grooming (e.g. Moore, 1986; Pleskacheva, 1996). The subjects were housed in groups of 3 and 4 in wire mesh cages (44 cm × 28 cm × 28 cm) at the Warsaw University Psychology Department animal colony, under standard laboratory conditions (12-light:12-dark cycle) on commercial rat chow diet and water ad libitum. The animals were tested at circa 83 days of age (Komorowska and Pisula, 2003). 2.1. Apparatus The testing apparatus consisted of an illuminated rectangular open arena (120 cm × 150 cm) and two shelters made of transparent glass, positioned in the centre of the arena. The surface of the arena was covered with grey linoleum, light was of moderate intensity (30 lx as measured at the surface of the arena). 2.2. Procedure On the day of testing, each animal was transported into the experimental room in an opaque plastic container and left undisturbed for a 10 min period. Afterwards, the rat was picked up by the experimenter (person known to subjects) and positioned in the apparatus for a period of 15 min. (It was established during preliminary tests that 15 min were sufficient for subjects to greatly reduce their activity in the apparatus.) For standardization purposes each rat was placed directly into one of the shelters. Behaviour of the subject animals was videotaped with a VHS
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
289
camera (Panasonic) and scored from the videotaped recordings. The surface of the arena and the interior of the shelters were thoroughly washed with soapy water and dried between subjects. 2.3. Description of behavioural categories Of all pelage oriented activities, the stereotyped grooming bouts were chosen for analyses, that is, those bouts that incorporated the syntactic chain (see Section 1 for details). Pauses in grooming longer than 8 s in duration were treated as a cut-off point between separate bouts. All analyses were performed on the first three grooming bouts, since further bouts were too infrequent to warrant their statistical comparison. Two general categories of within-bout activities were identified. These were: (A) Rostral grooming (RG): All actions performed by the forepaws and directed at the snout, face and head areas, including the ears. Within the category of rostral grooming a subcategory of preliminary rostral grooming (PRG) was identified. This subcategory incorporated forepaw licking and the wiping of the snout area. (B) Caudal grooming (CG): Pelage cleaning of the trunk and the extremities. This category incorporated the actions of licking, scratching and/or nibbling the fur of the belly, the flanks and the back, the anogenital area and the hind paws as well as the furry parts of the forelimbs (arm and/or shoulder). Within the general category of caudal grooming a subcategory of lateral caudal grooming (LCG) was identified that involved the licking, rubbing and/or nibbling of the flanks and the dorsal parts of the trunk. Each bout was divided into two segments (S1 and S2), with the syntactic chain marking the end of S1 and the beginning of S2 (Fig. 1). Three parameters pertaining to the S1 segment of each bout were recorded. These were: (1) total duration (measured as the number of 3 s intervals), (2) duration of uninterrupted preliminary rostral grooming (measured as the number of consecutive 3 s intervals), (3) presence of caudal grooming (probability of occurrence). Similarly, two bout parameters pertaining to the S2 segment were recorded. These were: (1) total duration (measured as the number of 3 s intervals),
Fig. 1. A schematic representation of a stereotyped grooming bout. The initial section (terminated by the syntactic chain) represents the S1 segment of the bout, the end section represents the S2 segment of the bout.
(2) duration of uninterrupted lateral caudal grooming (measured as the number of consecutive 3 s intervals). If more than one episode of lateral caudal grooming was present within a given bout, the longest episode was used for comparisons. 2.4. Statistical analyses Sample variances lacked homogeneity, therefore, the non-parametric Friedman analysis of variance was used for statistical comparisons. Subsequent pairwise comparisons were performed using the Wilcoxon matched-pair signed-ranks test (using a P < 0.05 cut-off).
3. Results There were significant differences in the duration of consecutive bouts for both S1 (χr2 (2) = 14.63, P < 0.001), and S2 (χr2 (2) = 13.63, P < 0.001, Fig. 2). Pairwise comparisons for both S1 and S2 revealed differences between bouts one and two and between bouts one and three, but not between bouts two and three. Between-bout differences were present in the duration of uninterrupted preliminary rostral grooming (χr2 (2) = 13.3, P < 0.001, Fig. 3A). PRG was significantly shorter for bout one than for bout two, it was also shorter for bout one than for bout three. No significant difference was present between-bout two and bout three. The duration of lateral caudal grooming changed in the similar manner (χr2 (2) = 20.83, P < 0.001, Fig. 3B). It was the shortest in the context of the first bout and increased in its duration for subsequent bouts. All pairwise comparisons revealed significant differences between bouts.
290
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
Fig. 2. Duration of segment S1 (left) and segment S2 (right) of consecutive grooming bouts. Data represent mean ± S.E.M.
The probability of caudal grooming within the S1 segment of a bout differed between consecutive bouts (χr2 (2) = 20.83, P < 0.05, Fig. 3C). Caudally directed intrusions into S1 were significantly less likely within bout one than within bout two and significantly less likely within bout one than within bout three. No significant difference was revealed between bouts two and three.
4. Discussion The duration of the S1 segment differed between consecutive bouts. The S1 of the initial bout was the shortest, while the S1 of the third bout was the longest. The increase in the duration of segment 1 appeared to arise as a result of two co-occurring events: the appearance of caudal grooming in the S1 segment of
Fig. 3. Micro-characteristics of the consecutive grooming bouts: duration of preliminary rostral grooming (A), duration of uninterrupted caudal grooming (B), presence of caudal grooming within the S1 segment of a bout (C). All data represent mean ± S.E.M.
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
the later occurring bouts and a prolonged period of preliminary rostral grooming within the framework of those bouts. The observed differences in bout characteristics shed light on the regulatory mechanisms that are involved in the control of grooming sequences. Bouts of novelty-induced grooming are believed to be stereotyped, with component grooming acts progressing in an orderly snout-face-head-trunk sequence (e.g. Richmond and Sachs, 1980; Sachs, 1988; Spruijt and Gispen, 1983). Our data do not fully support this general assumption. It is true that the first grooming bouts followed the rostro-caudal progression: such bouts started rostrally and proceeded smoothly past the rostro-caudal threshold of the syntactic chain. However, in the context of later bouts, the presence of caudal grooming prior to the occurrence of the syntactic chain indicates a disruption of the rostro-caudal pattern. This suggests that the rostro-caudal order of grooming actions may be less typical than is generally acknowledged. The presence of ‘disorganized’ bouts within the trial can not be interpreted as an inconsequential or incidental deviation from the rostro-caudal sequence. On the contrary, the fact that the disorganized bouts tended to follow the rostro-caudally organized bouts indicates an orderly transition between both types of bouts. It appears that the differences between both bout types in their sequential patterning may reflect a shift in the internal state of the grooming animal. Various lines of evidence point to such shift as being a reflection of a decrease in the level of stress. Neurophysiological studies demonstrate that the exposure to novelty triggers the stress response (e.g. Armario et al., 1986; De Boer et al., 1990; Pfister, 1979). They also indicate that the intensity of this response depends on the intensity of the stressor (Armario et al., 1986; De Boer et al., 1990) and that the habituation to novelty reduces the neurophysiological stress (De Boer et al., 1990; Kant et al., 1984; van Erp et al., 1994; Roth and Katz, 1979). On this basis, the initial conditions of the present study can be assessed as moderately stressful with subsequent attenuation of the stress response taking place during the course of the trial. Many behavioural variables are known for their low reliability as markers of stress and stress-related states (e.g. Rodgers, 1997). However, one type of risk assessment behaviour: the stretched-attend posture (SAP) stands out as a robust correlate of physiological stress.
291
For example, high frequency of SAP is associated with high corticosterone levels and elevated blood pressure (Rodgers, 1997; Dielenberg and McGregor, 2001; Rodgers et al., 1999) and it can be reduced by repeated exposures to novelty (Kaesermann, 1986; Choleris et al., 2001). Interestingly, high SAP scores have also been shown to immediately precede initial but not later occurring bouts of grooming (Komorowska and Pisula, 2003).This last result demonstrates that the first bouts of grooming performed in a novel arena are associated with higher physiological stress than the subsequent bouts. Overall, the available evidence suggests that the difference in the organizational pattern of consecutive bouts (their progressive disorganization) may be due, at least in part, to a gradual dearousal from stress. The attribution of bout disorganization to a decrease in the level of stress provides a clue as to why the abundance of non-rostro-caudal bouts has not been adequately acknowledged. Typical experimental conditions involve manipulations on the subject animals’ direct environment immediately prior to the recording of their behavioural activity. Evidence demonstrates that even those procedures that are believed to be relatively benign, for example the opening of the cages or the shuffling around of the subjects, cause noticeable elevation of the neurophysiological measures of stress (Rollin, 1998). It may be that under typical experimental conditions, the triggering of stress promotes the rostro-caudal organization of the grooming bouts. If low stress conditions are indeed required to promote bout disorganization, no disorganized bouts will be present under the usual experimental settings. In agreement with this interpretation, and contrary to the typical findings, Bolles (1960) observed that bouts of grooming emitted by rats in the undisturbed setting of home cages did not progress rostro-caudally. The second finding of the present study relates to the changes in the duration of the two constituent segments of consecutive bouts. Both segments were the shortest for the first bout and gradually increased in length for the following bouts. The fact that the increase in the duration of one segment was coupled with a simultaneous increase in the duration of the other segment means that entire bouts were increasing in their length over the course of the trial. Gradual prolongation of bouts has also been reported for other novel settings (in mice: Choleris et al., 2001 and
292
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
Fentress, 1972; in rats: Horvath et al., 1971; Hughes, 1991 and Pleskacheva, 1996) and may be related to the decrease in the level of stress over the course of novelty exposure (Choleris et al., 2001; Komorowska and Pisula, 2003; Pleskacheva, 1996). Interestingly, within the context of the present study, not only bouts themselves but also two component activities within those bouts changed in their overall duration. Both components: the preliminary rostral grooming and the lateral caudal grooming were emitted in the longest uninterrupted strings of activity within the final bouts. The pattern of changes observed for the two component activities suggests the involvement of two mechanisms in the shaping of these activities. A mechanism of activation was responsible for the initiation of a particular action, and a mechanism of continuation allowed the given action to be emitted for a period of time before it was replaced with another component action. Clearly, since both PRG and LCG were present in all analysed bouts, the mechanism of activation was operating unimpeded within the framework of all bouts. However, a relatively short duration of the two component actions within the first bouts suggests that the process of continuation was disturbed in the context of those bouts, but not in subsequent bouts. This dichotomy suggests that the two mechanisms (activation and continuation) should be viewed as operating relatively independently of one another. The results also point to another organizational property that operates on grooming behaviour. Since both PRG and LCG could be lengthened or shortened within the broader framework of individual grooming bouts, they should be treated as relatively independent units of those bouts. However, and despite their relative independence, the two activities were still part and parcel of the structurally diverse and fluidly executed sequences of motor actions. This suggests another mechanism—this mechanism is responsible for the integration of the grooming activities into uninterrupted and temporally discrete pelage cleaning sequences. It appears that grooming bouts constitute an end product of interplay between processes of activation and integration of constituent grooming activities. Research on the early development of face grooming in laboratory rodents lends support to the above interpretation. It points to various face directed strokes as loosely structured during the initial stages of devel-
opment. Originally, such actions tend to be performed separately and independently of each other (Fentress, 1972; Golani and Fentress, 1985). Only later in development, does this independence disappear and the unitary motor patterns become integrated into longer sequences of actions (Golani and Fentress, 1985). The fact that the mechanisms of activation and integration of component actions have been identified both for mature expression of novelty-triggered grooming and for the development of pelage care highlights the importance of both mechanisms in the organization of the grooming behaviour. An initial independence of motor actions and the subsequent integration of those component-actions, have also been observed during the development of several other naturally occurring behaviours (e.g. righting in rodents and marsupials: Pellis et al., 1992 and Pellis et al., 1991; locomotor co-ordination in rodents: Bekoff and Trainer, 1979).The interesting parallel between our findings regarding novelty-induced grooming and the findings regarding other patterned behaviours suggests that the interplay between the activation and the integration of the unitary component actions is not limited to pelage cleaning. This interplay may represent a fundamental principle that guides the production of patterned behaviour. In conclusion, the present results reveal that a number of regulatory processes are involved in the expression of novelty-induced grooming. Grooming starts rostro-caudally but becomes progressively disorganized as the trial continues—a pattern that may be due to a shift in the motivational state of the grooming animal, possibly a gradual dearousal from stress. The differences in the micro-characteristics of consecutive grooming bouts suggest the involvement of specific endogenous mechanisms in the shaping of those bouts. It appears that at least some component activities should be considered as units whose activation and continuation are controlled independently within the framework of individual grooming bouts. A mechanism of integration appears to be necessary to bind those component actions into cohesive, uninterrupted and protracted sequences of pelage cleaning. The organizational principles identified in the context of novelty-induced grooming may, in fact, be representative of fundamental organizational principles that govern co-ordinated behaviour in general.
J. Komorowska, S.M. Pellis / Behavioural Processes 67 (2004) 287–293
References Armario, A., Montero, J.L., Balasch, J., 1986. Sensitivity of corticosterone and some metabolic variables to graded levels of low intensity stresses in adult male rats. Physiol. Behav. 37, 559–561. Bekoff, A., Trainer, W., 1979. The development of interlimb co-ordination during swimming in postnatal rats. J. Exp. Biol. 83, 1–11. Berridge, K.C., 1989a. Progressive degradation of serial grooming chains by descending decerebration. Behav. Brain Res. 33, 241– 253. Berridge, K.C., 1989b. Substantia nigra 6-OHDA lesions mimic stratopallidal disruption of syntactic grooming chains: a neural systems analysis of sequence control. Psychobiology 17, 377– 385. Berridge, K.C., Fentress, J.C., 1986. Contextual control of trigeminal sensorimotor function. J. Neurosci. 9, 325–362. Berridge, K.C., Fentress, J.C., 1987. Disruption of natural grooming chains after stratopallidal lesions. Psychobiology 15, 336–342. Bolles, C., 1960. Grooming behaviour in the rat. J. Comp. Physiol. Psychol. 53, 306–310. Choleris, E., Thomas, A.W., Kavaliers, M., Prato, F.S., 2001. A detailed ethological analysis of the mouse open field test: effects of diazepam, chloriazepoxide and an extremely low frequency pulsed magnetic field. Neurosci. Biobehav. Rev. 25, 235– 260. De Boer, S.F., Koopmans, S.J., Slangen, J.L., van der Gugten, J., 1990. Plasma catecholamine, cortocosterone and glucose responses to repeated stress in rats: effect of interstressor interval length. Physiol. Behav. 47, 1117–1124. Dielenberg, R.A., McGregor, I.S., 2001. Defensive behaviour in rats towards predatory odors: a review. Neurosci. Biobehav. Rev. 25, 597–609. Fentress, J.C., 1972. Development and patterning of movement sequences in inbred mice. In: Kiger, J. (Ed.), The Biology of Behaviour. Oregon State University Press, Oregon, pp. 83– 132. Gallistel, C.R., 1980. The Organization of Action: A New Synthesis. Erlbaum Associates, Hillsdale, 432 pp. Golani, I., Fentress, J.C., 1985. Early ontogeny of face grooming in mice. Dev. Psychobiol. 18, 529–544. Grillner, S., 1975. Locomotion in vertebrates: central mechanisms and reflex interactions. Physiol. Rev. 55, 274–385. Horvath, T., Kirby, H.W., Smith, A.A., 1971. Rat’s heart rate and grooming activity in the open field. J. Comp. Physiol. Psychol. 73, 449–453. Hughes, R.N., 1991. The role of self- and other-animal-produced odors, in rats’ preferences for novelty, in an exploration box. Psychobiology 19, 168–174.
293
Kaesermann, H.P., 1986. Stretched attend posture, a non-social form of ambivalence, is sensitive to a conflict-reducing drug action. Psychopharmacology 89, 31–37. Kant, G., Eggleston, T., Landman-Roberts, L., Kenion, C., Driver, G., Meyerhoff, J., 1984. Habituation to repeated stress is stressor specific. Pharmacol. Biochem. Behav. 22, 631–634. Komorowska, J., Pisula, W., 2003. Does changing levels of stress affect the characteristics of grooming behaviour in rats? Int. J. Comp. Psychol. 16, 237–246. Moore, C.L., 1986. Sex differences in self-grooming of rats: effects of gonadal hormones and context. Dev. Psychobiol. 17, 243– 253. Pellis, S.M., Pellis, V.C., Nelson, J.E., 1992. The development of righting reflexes in the pouch young of the marsupial Dasyurus hallucatus. Dev. Psychobiol. 25, 105–125. Pellis, V.C., Pellis, S.M., Teitelbaum, P., 1991. A descriptive analysis of the postnatal development of contact-righting in rats. Dev. Psychobiol. 24, 237–263. Pfister, P., 1979. The glucocorticosterone response to novelty as a psychological stressor. Physiol. Behav. 23, 649–652. Pleskacheva, M.G., 1996. Temporal characteristics of grooming in an open field in two strains of rats. Int. J. Comp. Psychol. 9, 105–117. Rodgers, R., 1997. Animal models of ‘anxiety’: where next? Behav. Pharmacol. 8, 477–496. Rodgers, R.J., Haller, J., Holmes, A., Halasz, J., Walton, T.J., Brain, P.F., 1999. Corticosteron response to the plus-maze: high correlation with risk assessment in rats and mice. Physiol. Behav. 68, 47–53. Richmond, G., Sachs, B.D., 1980. Grooming in Norway rats: the development and adult expression of a complex motor pattern. Behaviour 75, 82–96. Rollin, B.E., 1998. The Unheeded Cry: Animal Consciousness, Animal Pain, and Science. Iowa State University Press, Ames, 330 pp. Roth, K., Katz, R., 1979. Stress, behavioural arousal, and open field activity – a reexamination of emotionality in the Rat. Neurosci. Biobehav. Rev. 3, 247–263. Sachs, B.D., 1988. The development of grooming and its expression in adult animals. Ann. N.Y. Acad. Sci. 525, 1–17. Spruijt, B., Gispen, W., 1983. Behavioural sequences as an easy quantifiable parameter in experimental studies. Physiol. Behav. 32, 707–710. Spruijt, B., van Hooff, J., Gispen, W., 1992. The ethology and neurobiology of grooming behaviour. Physiol. Rev. 72, 825– 852. van Erp, A., Kruk, M., Meelis, W., Willekens-Bramer, D., 1994. Effect of environmental stressors on time course, variability and form of self-grooming in the rat: handling, social contact, defeat, novelty, restraint and fur moistening. Behav. Brain Res. 65, 47–55.