Novelty affects paw preference performance in adult mice

Animal Behaviour 80 (2010) 51e57

Contents lists available at ScienceDirect

Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav

Novelty affects paw preference performance in adult mice Anderson Ribeiro-Carvalho, Yael Abreu-Villaça, Danielle Paes-Branco, Cláudio C. Filgueiras, Alex C. Manhães* Laboratório de Neurofisiologia, Departamento de Ciências Fisiológicas, Instituto de Biologia Roberto Alcantara Gomes, Centro Biomédico, Universidade do Estado do Rio de Janeiro

a r t i c l e i n f o Article history: Received 19 January 2010 Initial acceptance 11 February 2010 Final acceptance 19 March 2010 Available online 30 April 2010 MS. number: A10-00040 Keywords: brain asymmetry consistency of laterality habituation pawedness

The hemispheres are asymmetrically involved in the reaction to stressful situations. In this sense, it is possible to speculate that the asymmetrical activation of the hemispheres, as a result of the exposure to a novel situation, may affect behavioural lateralization. We tested the hypothesis that novelty affects performance in a paw preference task in 37 habituated (HAB) and 37 control (CT) adult male Swiss mice. For 4 days prior to the first testing session, HAB mice were placed in the testing box daily. After the fourth session, animals were deprived of food for 24 h. On the 5th day, food pellets were placed inside a feeding tube and animals were allowed to make 25 successful retrievals of food pellets. The testing procedure was repeated 4 days later. CT mice were not submitted to the habituation sessions. A significant sidedependent difference in consistency of laterality was observed between groups in the first session: all (100%) right-pawed CT mice used their right paw to make their first successful retrieval of food in the first testing session, while only 61% of left-pawed mice used their left paw. The same pattern was observed when the first five retrievals were considered: 100% right-pawed CT mice and 72% left-pawed CT mice were consistent. No differences were observed in the HAB group: in both side-preference subgroups, 88% of the animals showed consistent laterality. These results indicate that behavioural lateralization in paw preference is affected by the novelty of the testing situation in a side-dependent manner. Ó 2010 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

One of the most prominent features of behavioural lateralization in humans is the fact that the vast majority (approximately 90%) of the population consistently uses the right hand in tasks such as writing and drawing (Annett 2002). In spite of the fact that a considerable effort has been made in trying to assess the neurobiological basis of this remarkable behavioural trait, a comprehensive description of the mechanisms that determine human hand preference in particular and behavioural lateralization in general is not yet possible. Currently, it is believed that behavioural lateralization is the observable expression of functional specializations of the brain hemispheres (Rogers 2009). In turn, the functional specializations that have been uncovered so far, such as those related to language and attention, are thought to reflect biochemical and/or structural differences between the hemispheres (Hellige et al. 1998; Mesulam 1999; Josse & Tzourio-Mazoyer 2004). In trying to explain how asymmetrical hemispheres coordinate their functions during the execution of lateralized behaviours, both

* Correspondence: A. C. Manhães, Laboratório de Neurofisiologia, Departamento de Ciências Fisiológicas, Instituto de Biologia Roberto Alcantara Gomes, Centro Biomédico, Universidade do Estado do Rio de Janeiro, Av. Prof. Manoel de Abreu 444, 5 andar, Vila Isabel, Rio de Janeiro, RJ 20550-170, Brazil. E-mail address: [email protected] (A.C. Manhães).

animal models and humans have been studied (Vallortigara & Rogers 2005). Interestingly, some studies have shown that patterns of hemispheric lateralization change with practise (Castellano et al. 1989), even over the course of a single experiment (Streitfeld 1985; Burton & Wilson 1990; Fagot & Vauclair 1991, 1994; Van Horn et al. 1998). Shifts from an initial right-hemisphere advantage in early trials towards left-hemisphere advantage in later ones have been described with auditory (Burton & Wilson 1990), visual (Streitfeld 1985) and tactile (Streitfeld 1985) stimuli. In a similar way, nonhuman primates tested in somatosensory discrimination tasks presented a reduction of their initial predominance of left hand preferences with practise (Ettlinger & Moffett 1964; Brown & Ettlinger 1983). In toads (Robins & Rogers 2006), the left visual hemifield is preferentially used to observe and attack complex prey stimuli identified as novel. This left visual hemifield advantage disappeared with repeated testing. Interestingly, in these animals, predatory responses directed at familiar prey are usually carried out using the right visual hemifield. A similar shift in preference has also been observed for lizards (Robins et al. 2005). These results support the idea that the right hemisphere is specialized for the processing of new tasks, having an important role in the initial phases of acquisition, while the left hemisphere plays a major role in tasks for which the individual already has some training. Therefore, right-to-left shifts should be

0003-3472/$38.00 Ó 2010 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.anbehav.2010.03.024

52

A. Ribeiro-Carvalho et al. / Animal Behaviour 80 (2010) 51e57

expected after some practise. The theory also points to the possibility that shifts in lateralization at the population level that accompany practise reflect changes in lateralization in a subset of individuals: those for whom an initial advantage of the right hemisphere was observed (Kittler et al. 1989). Given the natural limitations that are present in primate studies, the existence of behavioural lateralization and functional specialization in other species allows for the systematic study of factors that may be implicated in the expression of these traits (Corballis 2009). Skilled reaching, as exemplified by forelimb use, has been identified in many species (Rogers 2009). Of particular interest to the present study is the fact that, in some of these species, individuals show a marked and consistent preference towards the use of one of the hands/paws. Despite the fact that hand/paw preference in animal models may not constitute a clear homologue, or even an analogue, of human hand preference, it can be used to assess the mechanisms underlying behavioural lateralization, providing evidence that may be used in understanding human lateralization (Rogers 2009), including hand preference. One of the most frequently used tests to assess pawedness in rodents, developed by Collins (1968, 1969), intrinsically involves exposure to a novel situation. In this test, after a period of food deprivation, mice are removed from their home cage and individually placed in a small testing box where they are required to perform a reaching task in order to retrieve small quantities of food that are made available in a feeding tube. In spite of its frequent use as an investigative tool, little is known about the possible effects that the novel, and potentially anxiogenic, situation has on overall performance during the test, including parameters such as degree and consistency of lateralization. Several studies demonstrated that the Collins’s paw preference test has a high intertest consistency of results regarding magnitude and direction of lateralization (Schmidt et al. 1991; Manhães et al. 2003, 2005). Much higher than the intertest reliability observed for other frequently used tests of behavioural lateralization in rodents such as the free-swimming and tail-suspension tests (Schmidt et al. 1999; Filgueiras & Manhães 2004, 2005; Manhães et al. 2007). This high reliability would seem to indicate that the patterns of lateralization observed in the paw preference test are not affected by factors such as the novelty of the testing situation. However, there are some indications that this might not be the case. For instance, it has been demonstrated that practise increases the magnitude of lateralization in several tests that assess behavioural lateralization (Castellano et al. 1989), including the Collins’s test (Bulman-Fleming et al. 1997). These results suggest that a learning process may be relevant to consolidate lateralization in tests that are heavily dependent on the animal’s ability to acquire new behaviours while dealing with novel situations. Furthermore, evidence is available indicating that the hemispheres are asymmetrically involved in the rodent’s reaction to stressful situations (Krahe et al. 2002; Filgueiras et al. 2006). In this sense, it is possible to speculate that the asymmetrical activation of the hemispheres as a result of the exposure to a novel situation may affect the initial selection of the paw to be used in retrieving the food in the Collins’s paw preference test and, as a result of a learning process, this initial choice would be readily consolidated, leading to a stable lateralization. One way to ascertain the effects of novelty on paw preference performance is to compare animals that were previously habituated to their testing environment (Leussis & Bolivar 2006) with animals that were not exposed to that environment prior to the paw preference task. Therefore, in the present study, we analysed the effects of novelty on performance and consistency of laterality in a paw preference task using a test adapted (Schmidt et al. 1991; Manhães et al. 1993, 2003, 2005) from a procedure initially described by Collins (1968, 1969).

METHODS Subjects Subjects were 74 (21 litters) adult Swiss male mice that were randomly assigned into two groups: habituated (HAB: N ¼ 37) and control (CT: N ¼ 37). All mice were bred and maintained in our laboratory in the same conditions without interventions during the first 21 postnatal days. The animals were kept in a temperaturecontrolled room on a 12 h light:dark cycle. Access to food and water was ad libitum (except when otherwise specified). All experiments were conducted in accordance with the declaration of Helsinki and with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health (U.S.A.). The experiments were conducted to minimize the number of animals used and the suffering caused by the procedures used in the present study. Testing Procedure Paw preference test began at adulthood (3 months). For the habituated group, the first test was preceded by a habituation period of 4 days in which the animals were individually placed in the paw preference testing box (7.5  7.5  14.0 cm, W  L  H) for 15 min at a time, once a day. The box had a transparent front wall with a cylindrical feeding tube attached to it in an equidistant position from the two side walls (Schmidt et al. 1991; Manhães et al. 1993, 2003, 2005). During this period, the animals had unrestricted access to food in their cages but no food was available in the feeding tube. After leaving the box on the last day of the habituation procedure, animals were deprived of food for 24 h. On the 5th day, animals were once more placed in the testing box to begin the first paw preference testing session. This time, however, the feeding tube was kept empty only during the first 5 min of the testing session, a period that allowed the initial exploratory activity to subside. After this period, animals were required to perform a reaching task in order to retrieve food pellets (4e5 mm in diameter) that were made available in the feeding tube. The food pellets were the same that were regularly used to feed the animals, so that all mice were used to the smell, taste and texture of the pellets. Given the placement of the feeding tube, the food could be equally accessed by the right or the left paw. Therefore, left or right paw reaches could be easily observed. Every behavioural session consisted of 25 reaches for food. Four days later, animals were again deprived of food for 24 h. Then, animals were placed in the testing box to begin the second testing session, which followed the same procedure described for the first session. In the second session, the experimenters were blind as to the scores previously obtained in the first session. Control (CT) mice were not subjected to the 4-day habituation period prior to food deprivation, and food was made available in the feeding tube immediately after the animals were placed in the testing box. All testing sessions were videotaped and the experimenter only started recording when the first pellet was placed in the feeding tube. These recordings were subsequently used to assess paw preference performance. Laterality We performed the analysis of laterality using procedures described in previous reports (Schmidt et al. 1991; Manhães et al. 1993, 2003, 2005). Briefly, we noted the paw that was used to remove the pellet from the feeding tube. For each animal, the percentage of left paw use (%L ¼ number of left paw reaches in both sessions  100/50) was used to evaluate side preference and the percentage of preferred paw use (%PP ¼ number of reaches with

A. Ribeiro-Carvalho et al. / Animal Behaviour 80 (2010) 51e57

Consistency of Laterality For left- or right-pawed mice, as indicated by the global analysis of laterality (results from both the first and second sessions considered together), if a given animal’s first reach in the first session was done with the same paw for which the global analysis of its laterality indicated preference, the animal would be considered consistent. Consistency of laterality was also assessed using the following parameters separately: first five reaches in the first session (three or more reaches with a given paw indicated the side preference); first reach in the second session; first five reaches in the second session. Time To analyse the performance of each mouse, in terms of the time it took to perform specific events in each behavioural session, we recorded each animal’s latency (in seconds): (1) to place its snout into the feeding tube (to verify that food was present in the feeding tube); (2) to make the first attempt (successful or not) to retrieve the food; (3) to make the first (R1), fifth (R5), 10th (R10) and last reach (R25). The chronometer was started when the food was placed in the feeding tube. The events were chosen based on previous experience using the paw preference testing apparatus (Schmidt et al. 1991; Manhães et al. 1993, 2003, 2005), which suggested that significant transitions in animals’ reaching performance would be observed at these points. However, because events were not evenly spaced along the testing session, we calculated the average times per reach within the following four intervals: (1) interval 1, which began when the animal made its first attempt to retrieve the food with a paw and ended when the first successful retrieval of the food was made; (2) interval 2, which began at the first reach and ended at the fifth reach; (3) interval 3, which began at the fifth reach and ended at the 10th reach; (4) interval 4, which began at the 10th reach and ended at the last attempt. Statistical Analyses We assessed the normality of the distributions of each variable using KolmogoroveSmirnov one-sample tests. We analysed differences between groups in the distributions of %L and %PP using ManneWhitney U tests, and differences between groups in the percentage of lateralized, left-pawed and consistent animals using chi-square tests or, if expected counts were less than 5, with Fisher’s exact tests. Time data are presented as means and SEs, which were calculated from raw data. Statistical analyses, however, were performed on log-transformed data because of heterogeneity of the variances and skewness of the distributions. We examined differences between groups in latency to perform each event using a repeated measures analysis of variance (rANOVA). Group (habituated or control) was used as the between-subjects factor. Session (first and second) and event (locate, attempt, R1, R5, R10 and R25) were used as within-subjects factors. Differences between groups in average time per reach were also examined using rANOVA. Group (habituated or control) was used as the between-subjects factor. Session (1, 2) and interval (1e4) were used as withinsubjects factors. Lower-order ANOVAs were used as post hoc tests.

Effects were considered significant when P < 0.05 (two tailed). For interactions at P < 0.10 (two tailed), we also examined whether lower-order main effects were detectable after subdivision of the interactive variables (Snedecor & Cochran 1967). For simplicity, we report results based only on the averaged univariate F tests. The univariate approach is considered more powerful than the multivariate criteria (Huynh & Feldt 1976). However, each univariate test requires that the variances of all transformed variables for an effect must be equal and their covariances must be zero (Huynh & Feldt 1976). Therefore, we estimated the extent to which the covariance matrices deviated from sphericity using the Mauchly’s test. Whenever the sphericity assumption was violated, we used the GreenhouseeGeisser correction, which adjusts the degrees of freedom, in order to avoid type I errors. RESULTS Laterality The distributions of the percentage of preferred paw use (%PP) of CT and HAB mice are shown in Fig. 1a. Both distributions were similarly J-shaped: most animals were highly lateralized (independent of side preference). There were no significant differences in the shape of the distributions of %PP between groups

80

(a)

70

More lateralized

Less lateralized

60 50 40 30 20 Relative frequency (%)

the preferred paw in both sessions  100/50) was used to evaluate the magnitude of laterality independent of side preference. To evaluate whether a given animal had a significant side preference (either leftward or rightward), a binomial Z score was calculated based on the number of left paw reaches. Mice with jZj values higher than 1.96 were considered lateralized and those having jZj values lower than 1.96 were considered nonlateralized.

53

10 0 50

60

70

80

90

100

% PP 50 45 40

(b) More right-pawed

More left-pawed

35 30 25 20 15 10 5 0

20

40

60

80

100

%L Figure 1. Frequency distributions of (a) preferred paw use (%PP) and (b) left-paw use (%L) in adult male Swiss mice. ,: control; G: habituated.

A. Ribeiro-Carvalho et al. / Animal Behaviour 80 (2010) 51e57

(ManneWhitney U test: U ¼ 675, N1 ¼ N2 ¼ 37, P ¼ 0.92). Furthermore, the percentage of significantly lateralized animals (those classified as either left- or right-pawed) was significantly different from chance level in both groups: 89.2% in the CT group (binomial Z score: Z ¼ 4.8, P < 0.001) and 91.9% in the HAB group (Z ¼ 5.1, P < 0.001). No significant differences were found between groups (chi-square test: c21 ¼ 0.2, P ¼ 0.69) in the percentage of lateralized animals. The distributions of the percentage of left paw use (%L) of CT and HAB mice are shown in Fig. 1b. Both distributions were similarly U-shaped: the number of animals that were considered left-pawed was similar to that of right-pawed ones. There were no significant differences in the shape of the distributions of %L between groups (ManneWhitney U test: U ¼ 679, N1 ¼ N2 ¼ 37, P ¼ 0.95). The percentage of animals classified as right-pawed, left-pawed or nonlateralized is presented in Table 1. No significant differences were found between groups (chi-square test: c22 ¼ 0.3, P ¼ 0.86). Consistency of Laterality The percentage of lateralized animals in the HAB group did not vary from the first (86%) to the second (89%) testing session. The percentage of lateralized animals in the CT group increased from the first (73%) to the second (89%) session, however, the difference only approached significance (chi-square test: c21 ¼ 3.2, P ¼ 0.07). Most animals, in both groups, presented the same side preference in both testing sessions: 89% in the HAB group and 76% in the CT group. Consistency of laterality varied between subgroups when global (first and second sessions) side preference was taken into account (ambidextrous animals were not used at this point because of the small sample sizes; HAB: N ¼ 3; CT: N ¼ 4). In the CT group, 100% (N ¼ 15) of right-pawed animals used the right paw to make their first successful reach in the first testing session (Fig. 2, Table 2). As for the left-pawed animals in this group, only 61% (N ¼ 11) used the left paw to make their first successful reach. The difference between left- and right-pawed groups was highly significant (Fisher’s exact test: P ¼ 0.009). Expanding the analysis to include the first five reaches (Table 2), a significant difference (Fisher’s exact test: P ¼ 0.049) between side-preference groups was still observed in CT animals: 100% (N ¼ 15) of right-pawed animals used their right paw three or more times and only 72% of left-pawed animals used their left paw three or more times. No differences between side-preference subgroups of the habituated animals were observed regarding consistency of laterality in the first testing session (Table 2). Furthermore, no differences in consistency of laterality between side-preference subgroups of both the CT and HAB animals were observed in the second testing session (Table 2). Time Latencies to each one of the events are presented in Table 3. The rANOVA indicated a significant effect of session (F1,68 ¼ 208.1, P < 0.001) as well as significant sessionevent (F2.0,138.9 ¼ 57.4, P < 0.001) and session  event  group (F2.0,138.9 ¼ 2.5, P ¼ 0.086) interactions. Considering the session effect (latencies in the second session were shorter) and interactions, subsequent analyses were performed separately for each session. In the first testing session, Table 1 Global side preference (1st and 2nd testing sessions)

Control Habituated

Nonlateralized

Left-pawed

Right-pawed

10.8% (N¼4) 8.1% (N¼3)

40.5% (N¼15) 45.9% (N¼17)

48.6% (N¼18) 45.9% (N¼17)

**

100 90 Relative frequency (%)

54

80 70 60 50 40 30 20 10 0

Left

Right Control

Left Right Habituated

Figure 2. Consistency of laterality in left- and right-pawed mice based on the paw (left or right) that mice used in their first successful reach in the first session that corresponded to the paw for which a global preference (1st þ 2nd sessions) was demonstrated. **P < 0.01. ,: inconsistent; G: consistent.

the lower-order rANOVA indicated a significant effect of group (F1,68 ¼ 12.0, P ¼ 0.001) as well as a significant group  event interaction (F2.3,159.6 ¼ 8.2, P < 0.001). These results can be explained by the fact that HAB mice presented lower latencies for the attempt, R1, R5 and R10 events when compared to CT mice, while no differences were observed regarding latency to locate the pellet and to make the last attempt (Table 3). In the second session, a significant effect of group (rANOVA: F1,68 ¼ 9.3, P ¼ 0.003) and a significant group  event interaction (rANOVA: F2.1,141.6 ¼ 3.4, P ¼ 0.034) were also observed. These results can be explained by the fact that HAB mice presented lower latencies for the locate, attempt, R1, R5 and R10 events when compared to CT mice, while no differences were observed regarding latency to the last attempt (Table 3). Regarding the average time per reach (Table 4), in the first session the rANOVA indicated significant effect of session (F1,68 ¼ 241.9, P < 0.001) and significant session  interval interaction (F2.1,144.8 ¼ 126.5, P < 0.001). Considering the session effect (averages were lower in the second session) and interaction, subsequent analyses were performed separately for each session. In the first session, a lower-order rANOVA indicated a significant effect of interval (F2.2,147.8 ¼ 229.4, P < 0.001) and a significant group  interval interaction (F2.2,147.8 ¼ 7.2, P ¼ 0.001). The group effect only approached significance (F1,68 ¼ 2.9, P ¼ 0.094). These results can be explained by the fact that, for both groups, performance improved from the first to the third interval, remaining relatively stable thereafter. However, the HAB advantage in the first interval was lost by the fourth interval. In fact, CT mice became faster than HAB mice in this last interval. In the second session, significant effects of group (rANOVA: F1,68 ¼ 5.9, P ¼ 0.018) and interval (F1.8,124.8 ¼ 4.5, P ¼ 0.015) were observed. A significant group  interval interaction (F1.8,124.8 ¼ 5.6, P ¼ 0.006) was also present. These results can be explained by the fact that the CT group still presented a marked improvement in performance between the first and third intervals that was no longer present in the HAB group, which in fact had a more stable, and better, performance throughout the session. Neither laterality nor consistency of laterality affected time performance. DISCUSSION Corroborating previous findings (Schmidt et al. 1991; Manhães et al. 1993, 2003, 2005), the majority of Swiss mice were

A. Ribeiro-Carvalho et al. / Animal Behaviour 80 (2010) 51e57

55

Table 2 Consistency of laterality Global side preference (1st & 2nd sessions)

1st Testing session

2nd Testing session

R1

R1eR5

R1

R1eR5

Control

Right-pawed Left-pawed

100% (N¼15)** 61% (N¼11)

100% (N¼15)* 72% (N¼13)

87% (N¼13) 94% (N¼17)

100% (N¼15) 100% (N¼18)

Habituated

Right-pawed Left-pawed

88% (N¼15) 88% (N¼15)

82% (N¼14) 94% (N¼16)

94% (N¼16) 88% (N¼15)

94% (N¼16) 94% (N¼16)

Animals classified according to their global side preference (1st and 2nd testing sessions). R1: paw used in the first retrieval; R1eR5: laterality defined according the first five retrievals in a session. *P < 0.05; **P < 0.01.

significantly lateralized, meaning that most animals had a marked side preference. Our current set of results also confirms the fact that Swiss mice have no population bias favouring one paw: the percentage of left-pawed animals was similar to that of rightpawed ones. In addition, the habituation procedure had no effect on the magnitude or direction of the behavioural lateralization since no differences were found between groups regarding these parameters. As for consistency of laterality, the CT group presented a marked side-dependent effect in the first testing session. All right-pawed mice (results of both testing sessions considered together) used their right paw to make their first food retrieval from inside the feeding tube. Conversely, only 61% of left-pawed animals used their left paw for their first retrieval. Even when the analysis was expanded to include a given animal’s first five retrievals, this side-dependent difference was present. The side-dependent effect was no longer present in the second testing session. Interestingly, most animals in the CT group did not switch their side preference between sessions. Taken together, these results suggest that whatever effect is interfering with side-preference consistency is more intense during the initial attempts, successful or not, to retrieve the food pellets. Differences in consistency between side-preference subgroups of CT mice could be the result of interactions between asymmetries controlling the learning component of the task and asymmetries controlling the paw preference. However, the absence of sidepreference effects on consistency of laterality in HAB mice renders this possibility unlikely. This lack of effect on consistency in the HAB mice points to an interesting possibility. Differences in consistency between side-preference subgroups of CT might have occurred because brain asymmetries associated with behaviour in novel situations interfered with paw selection. Mounting evidence indicates that, for many species, the hemispheres are asymmetrically used when dealing with new or threatening situations (Vallortigara & Rogers 2005). The diversity of species involved is such that it has been proposed that this trait has appeared fairly early during evolution (MacNeilage et al. 2009). More to the point, recent studies indicate that shifts in lateralization do occur as animals habituate to novel stimuli/environments. For example, both toads (Robins & Rogers 2006) and lizards (Robins et al. 2005)

tend to change the visual hemifield used to observe and strike novel prey as they habituate to it upon repeated exposure. Our current set of results is in line with these previous findings: The novelty of our testing set-up did affect the initial pattern of paw preference. As the test progressed, animals habituated to their environment and the sensoryemotor asymmetries determined subsequent behaviour. The neurobiological mechanisms underlying the asymmetrical involvement of the hemispheres in behavioural responses to novel situations have been under scrutiny for quite some time now and may help explain inconsistencies in CT mice. Several studies have demonstrated that mild stressor exposure activates dopaminecontaining neuronal systems projecting to the medial prefrontal cortex, nucleus accumbens and striatum (Carlson et al. 1991; Sullivan & Gratton 1998; van der Elst et al. 2005). Interestingly, some studies have demonstrated that this phenomenon is asymmetrical (Carlson et al. 1991, 1993). For example, a short period (15 min) of movement restraint causes a left to right dopaminergic activation in prefrontal cortex (Carlson et al. 1991). Of note, dopamine has been associated with pawedness in several studies (Schwarting et al. 1987; Cabib et al. 1995; Nielsen et al. 1997; Morice et al. 2005; Budilin et al. 2008). For instance, it has been demonstrated, in rats, that left-pawed animals have a higher concentration of dopamine in the left nucleus accumbens (Budilin et al. 2008). This pattern of association between paw preference and dopamine content in the hemispheres in general and in the nucleus accumbens in particular has also been observed in mice (Cabib et al. 1995). In fact, the higher degree of lateralization in the paw preference tests seems to be related to stronger asymmetries in dopamine content in the prefrontal cortex (Nielsen et al. 1997). As additional evidence that dopamine content affects lateralization, it has been shown that hyperdopaminergia directly impairs the degree of lateralization in mice lacking the dopamine transporter (DAT) gene (Morice et al. 2005). In order to test the association between asymmetries in dopamine levels and metabolism in the brain during exposure to the novel situations and the effects of these asymmetries on behavioural lateralization, a next step would be to study the content of this neurotransmitter and its metabolites as a function of paw preference in habituated and nonhabituated mice.

Table 3 Latency (in seconds) to locate, attempt to retrieve and reach for (R1-R25) food in each session Session

Group

Events Locate

Attempt

R1

R5

R10

R25

1st

Control (CT) Habituated (HAB) D (CTHAB)

354 428 7

41860*** 21162 207

56980** 37281 197

78488** 54981 235

91596* 68787 228

1281120 1147110 134

2nd

Control Habituated D (CTHAB)

315* 256 6

7116** 337 38

10122** 447 57

21032** 10912 100

29938** 17717 122

57858 51442 63

Data are presented as means  SE. D: difference; R1: 1st reach; R5: 5th reach; R10: 10th reach; R25: 25th (last) reach. Asterisks indicate pairwise comparisons within each session: *P < 0.05; **P < 0.01; ***P < 0.001.

56

A. Ribeiro-Carvalho et al. / Animal Behaviour 80 (2010) 51e57

References

Table 4 Average time per reach (in seconds) to retrieve food Session

Group

Interval 1

2

3

4

1st

Control (CT) Habituated (HAB) D (CTHAB)

53479** 33079 204

548 447 10

264 285 1

243* 313 6

2nd

Control Habituated D (CTHAB)

7019* 203 50

274** 162 11

182 141 4

192 222 4

Data presented as means  SE. D: difference. Asterisks indicate pairwise comparisons within each session: *P < 0.05; **P < 0.01; ***P < 0.001.

Regarding time data, our main finding was that both groups showed a marked improvement in performance between and within sessions, indicating that paw preference performance is associated with the acquisition, practise and retention of new skills. Overall, latencies were shorter for the HAB group, primarily because CT mice took considerably longer to make their first attempt (successful or not) than HAB ones. From then on, differences remained relatively stable. These observed differences are most likely related to the fact that animals had to focus their efforts on the task of retrieving the food, which comprises a complex set of movements (Alaverdashvili et al. 2008). As a result of their continued efforts, mice eventually adopted the body posture that allowed them to insert their paws into the feeding tube and make an attempt to grasp the pellet. Since CT mice were also dealing with a novel environment, a considerable fraction of the time during the initial moments of the first session was dedicated to exploring the testing box. The novelty effect was considerably attenuated in HAB mice, which would account for the observed group differences. This pattern of results suggests that novelty mainly affects the period within which animals are working out the mechanics of the task (e.g. inserting their paws into a feeding tube in order to retrieve the food). The highest gain in performance occurred from the first to the fifth retrieval. From the fifth to the 10th retrieval, gains in performance, albeit significant, were smaller, indicating that the first five retrievals were the most critical in learning the task. Conclusion The present study provides evidence that asymmetries controlling behaviour during novel experiences may affect, in a side-dependent manner, the expression of motor asymmetries during those same experiences. Interestingly, although side inconsistency was observed during a phase in which the animals were still learning the task, continued testing resulted in a stable preference pattern. This finding provides additional support for the hypothesis that behavioural lateralization is due to an intrinsic brain asymmetry, not to a learning process resulting from the performance of a successful attempt by a randomly chosen paw. Once the novelty effect subsides, after continued exposure to the testing situation, asymmetries related to motor performance will control behaviour, resulting in the stable pattern of lateralization that usually characterizes paw preference tests. Acknowledgments This work was supported by grants from Fundação Carlos Chagas Filho de Amparo a Pesquisa do Estado do Rio de Janeiro (FAPERJ) and Sub-Reitoria de Pós-graduação e Pesquisa da Universidade do Estado do Rio de Janeiro (SR2-UERJ). We thank Edson Oliveira for animal care.

Alaverdashvili, M., Leblond, H., Rossignol, S. & Whishaw, I. Q. 2008. Cineradiographic (video X-ray) analysis of skilled reaching in a single pellet reaching task provides insight into relative contribution of body, head, oral, and forelimb movement in rats. Behavioural Brain Research, 192, 232e247. Annett, M. 2002. Handedness and Brain Asymmetry. East Sussex: Psychology Press. Brown, J. V. & Ettlinger, G. 1983. Intermanual transfer of mirror-image discrimination by monkeys. Quarterly Journal of Experimental Psychology B, 35, 119e124. Budilin, S. Y., Midzyanovskaya, I. S., Shchegolevskii, N. V., Ioffe, M. E. & Bazyan, A. S. 2008. Asymmetry in dopamine levels in the nucleus accumbens and motor preference in rats. Neuroscience and Behavioral Physiology, 38, 991e994. Bulman-Fleming, M. B., Bryden, M. P. & Rogers, T. T. 1997. Mouse paw preference: effects of variations in testing protocol. Behavioural Brain Research, 86, 79e87. Burton, A. & Wilson, P. 1990. Hemisphere priming through practice in a musical chords task. Neuropsychologia, 28, 889e893. Cabib, S., D’Amato, F. R., Neveu, P. J., Deleplanque, B., Le Moal, M. & PuglisiAllegra, S. 1995. Paw preference and brain dopamine asymmetries. Neuroscience, 64, 427e432. Carlson, J. N., Fitzgerald, L. W., Keller, R. W. & Glick, S. D. 1991. Side and region dependent changes in dopamine activation with various durations of restraint stress. Brain Research, 550, 313e318. Carlson, J. N., Fitzgerald, L. W., Keller, R. W. Jr. & Glick, S. D. 1993. Lateralized changes in prefrontal cortical dopamine activity induced by controllable and uncontrollable stress in the rat. Brain Research, 630, 178e187. Castellano, M. A., Diaz-Palarea, M. D., Barroso, J. & Rodriguez, M. 1989. Behavioral lateralization in rats and dopaminergic system: individual and population laterality. Behavioral Neuroscience, 103, 46e53. Collins, R. L. 1968. On the inheritance of handedness. I. Laterality in inbred mice. Journal of Heredity, 59, 9e12. Collins, R. L. 1969. On the inheritance of handedness. II. Selection for sinistrality in mice. Journal of Heredity, 60, 117e119. Corballis, M. C. 2009. The evolution and genetics of cerebral asymmetry. Philosophical Transactions of the Royal Society B, 364, 867e879. van der Elst, M. C., Verheij, M. M., Roubos, E. W., Ellenbroek, B. A., Veening, J. G. & Cools, A. R. 2005. A single exposure to novelty differentially affects the accumbal dopaminergic system of apomorphine-susceptible and apomorphineunsusceptible rats. Life Sciences, 76, 1391e1406. Ettlinger, G. & Moffett, A. 1964. Lateral preferences in the monkey. Nature, 204, 606. Fagot, J. & Vauclair, J. 1991. Manual laterality in nonhuman primates: a distinction between handedness and manual specialization. Psychological Bulletin, 109, 76e89. Fagot, J. & Vauclair, J. 1994. Video-task assessment of stimulus novelty effects on hemispheric lateralization in baboons (Papio papio). Journal of Comparative Psychology, 108, 156e163. Filgueiras, C. C. & Manhães, A. C. 2004. Effects of callosal agenesis on rotational side preference of BALB/cCF mice in the free swimming test. Behavioural Brain Research, 155, 13e25. Filgueiras, C. C. & Manhães, A. C. 2005. Increased lateralization in rotational side preference in male mice rendered acallosal by prenatal gamma irradiation. Behavioural Brain Research, 162, 289e298. Filgueiras, C. C., Abreu-Villaça, Y., Krahe, T. E. & Manhães, A. C. 2006. Unilateral hemispherectomy at adulthood asymmetrically affects immobile behavior of male Swiss mice. Behavioral Brain Research, 172, 33e38. Hellige, J. B., Taylor, K. B., Lesmes, L. & Peterson, S. 1998. Relationships between brain morphology and behavioral measures of hemispheric asymmetry and interhemispheric interaction. Brain Cognition, 36, 158e192. Huynh, H. & Feldt, L. S. 1976. Estimation of BOX correction for degrees of freedom from sample data in randomized block and split-plot designs. Journal of Educational Statistics, 1, 69e82. Josse, G. & Tzourio-Mazoyer, N. 2004. Hemispheric specialization for language. Brain Research Reviews, 44, 1e12. Kittler, P., Turkewitz, G. & Goldberg, E. 1989. Shifts in hemispheric advantage during familiarization with complex visual patterns. Cortex, 25, 27e32. Krahe, T. E., Filgueiras, C. C. & Schmidt, S. L. 2002. Effects of rotational side preferences on immobile behavior of normal mice in the forced swimming test. Progress in Neuro-psychopharmacology & Biological Psychiatry, 26, 169e176. Leussis, M. P. & Bolivar, V. J. 2006. Habituation in rodents: a review of behavior, neurobiology, and genetics. Neuroscience & Biobehavioral Reviews, 30, 1045e1064. MacNeilage, P. F., Rogers, L. J. & Vallortigara, G. 2009. Origins of the left & right brain. Scientific American, 301, 60e67. Manhães, A. C., Schmidt, S. L. & Caparelli-Dáquer, E. M. 1993. Paw preference in mice with callosal defects induced by prenatal gamma irradiation. Brazilian Journal of Medical and Biological Research, 26, 1213e1218. Manhães, A. C., Krahe, T. E., Caparelli-Dáquer, E., Ribeiro-Carvalho, A., Schmidt, S. L. & Filgueiras, C. C. 2003. Neonatal transection of the corpus callosum affects paw preference lateralization of adult Swiss mice. Neuroscience Letters, 348, 69e72. Manhães, A. C., Schmidt, S. L. & Filgueiras, C. C. 2005. Callosal agenesis affects consistency of laterality in a paw preference task in BALB/cCF mice. Behavioural Brain Research, 159, 43e49.

A. Ribeiro-Carvalho et al. / Animal Behaviour 80 (2010) 51e57 Manhães, A. C., Abreu-Villaça, Y., Schmidt, S. L. & Filgueiras, C. C. 2007. Neonatal transection of the corpus callosum affects rotational side preference in adult Swiss mice. Neuroscience Letters, 415, 159e163. Mesulam, M. M. 1999. Spatial attention and neglect: parietal, frontal and cingulate contributions to the mental representation and attentional targeting of salient extrapersonal events. Philosophical Transactions of the Royal Society B, 354, 1325e1346. Morice, E., Denis, C., Macario, A., Giros, B. & Nosten-Bertrand, M. 2005. Constitutive hyperdopaminergia is functionally associated with reduced behavioral lateralization. Neuropsychopharmacology, 30, 575e581. Nielsen, D. M., Visker, K. E., Cunningham, M. J., Keller, R. W. Jr., Glick, S. D. & Carlson, J. N. 1997. Paw preference, rotation, and dopamine function in Collins HI and LO mouse strains. Physiology & Behavior, 61, 525e535. Robins, A. & Rogers, L. J. 2006. Complementary and lateralized forms of processing in Bufo marinus for novel and familiar prey. Neurobiology Learning & Memory, 86, 214e227. Robins, A., Chen, P., Beazley, L. D. & Dunlop, S. A. 2005. Lateralized predatory responses in the ornate dragon lizard (Ctenophorus ornatus). Neuroreport, 16, 849e852. Rogers, L. J. 2009. Hand and paw preferences in relation to the lateralized brain. Philosophical Transactions of the Royal Society B, 364, 943e954.

57

Schmidt, S. L., Manhães, A. C. & de Moraes, V. Z. 1991. The effects of total and partial callosal agenesis on the development of paw preference performance in the BALB/cCF mouse. Brain Research, 545, 123e130. Schmidt, S. L., Filgueiras, C. C. & Krahe, T. E. 1999. Effects of sex and laterality on the rotatory swimming behavior of normal mice. Physiology & Behavior, 65, 607e616. Schwarting, R., Nagel, J. A. & Huston, J. P. 1987. Asymmetries of brain dopamine metabolism related to conditioned paw usage in the rat. Brain Research, 417, 75e84. Snedecor, G. W. & Cochran, W. G. 1967. Statistical Methods, 6th edn. Ames: Iowa State University Press. Streitfeld, B. 1985. Hemispheric advantages and shifts of laterality differences in visual and tactual modalities. International Journal of Neuroscience, 25, 245e254. Sullivan, R. M. & Gratton, A. 1998. Relationships between stress-induced increases in medial prefrontal cortical dopamine and plasma corticosterone levels in rats: role of cerebral laterality. Neuroscience, 83, 81e91. Vallortigara, G. & Rogers, L. J. 2005. Survival with an asymmetrical brain: advantages and disadvantages of cerebral lateralization. Behavioral Brain Science, 28, 575e589. Van Horn, J. D., Gold, J. M., Esposito, G., Ostrem, J. L., Mattay, V., Weinberger, D. R. & Berman, K. F. 1998. Changing patterns of brain activation during maze learning. Brain Research, 793, 29e38.