Induction of Compulsive Checking
Two characteristics of compulsive checking - a preoccupation with the performance of the behavior and a reluctance to leave the place/object on which the behavior is focused - are said to be present if the subject meets three performance criteria [16
]: the subject returns to one or two spots in its territory excessively often, excessively rapidly, and visits excessively few other places before returning to the spot of interest. In a large open field, one of these spots is the rat's home base [16
] as defined by Eilam and Golani [19
]. Consequently, in the present study, the presence of compulsive checking was examined with reference to the home base. As found previously [16
], Figure shows that 10 injections of quinpirole induced compulsive checking of home base, according to the above 3 criteria. In particular:
Figure 1 Induction of compulsive checking as identified by formal performance criteria. Performance measures are in reference to the home base established by the rat on the tenth open field test shown here, and recognized as the locale with the longest total duration (more ...)
(1) As shown in Figure , quinpirole-treated rats revisited their home base almost 15 times more often than did saline-treated animals (141.6 ± 13.5 returns under quinpirole vs. 9.7 ± 1.6 returns under saline; t(22) = 9.7, p < .001). Moreover, even after adjusting for the total number of visits, returns to home base were excessive in quinpirole rats because the rate of returning to the home base was significantly higher compared to the rate of visits to other locales in the open field, and significantly greater than the rate of home base return in saline controls (Figure ). Specifically, quinpirole-treated rats made 406.7 ± 25.1 visits to 19.7 ± 0.9 different locales in the open field, and thus on the basis of a uniform frequency distribution their expected rate of return to any locale in the open field was 21.1 ± 1.5 returns per locale. However, the observed rate of revisits to the home base was almost 7-fold higher than would be expected. This number was also significantly higher than the corresponding ratio of observed-to-expected visits to home base in saline controls (6.8 ± 0.5 under quinpirole vs. 2.2 ± 0.1 under saline, t(22) = 9.3, p < .001; Figure ).
(2) As shown in Figure , the mean return time to the home base was 10-fold shorter in the quinpirole group than in the saline-treated rats (15.6 ± 2.4 s vs. 158.7 ± 20.4 s, t(22) = 7.0, p < .001). Moreover, return time to home base remained excessively rapid in quinpirole rats even when compared to their overall mean return time to places of visit in the open field (15.6 ± 2.4 s vs. 406.3 ± 35.8 s, t(11) = 10.9, p < .001), or when compared to saline controls using a normalized measure of return time, namely, home base return time normalized to overall return time (4.5 ± 1.2% of overall return time under quinpirole vs. 40.2 ± 6.9% under saline, t(22) = 5.1, p < .001).
(3) Finally, as shown in Figure , quinpirole-treated rats visited only a couple of places before returning to their home base, in contrast to saline controls which visited 4 times as many locales before re-entering home (2.0 ± 0.2 places under quinpirole vs. 7.9 ± 0.6 places under saline, t(22) = 9.5, p < .001).
Two additional criteria were proposed as required to identify in the rat compulsive checking behavior: the presence of a characteristic set of acts performed at the spot of interest, and a change in the pattern of checking behavior in response to a re-arrangement of the test environment (for a rationale for these criteria, see [16
]). Although motor acts were not scored in the present study, indirect evidence does suggest that quinpirole-treated rats met the criterion of ritual-like motor activity. Specifically, quinpirole rats spent significantly less time at the home base during each visit than did saline controls (9.7 ± 1.1 s per home visit under quinpirole vs.
308.2 ± 68.2 s under saline, t
(22) = 4.4, p
< .001), a finding associated previously with the differential display of ritual-like acts in quinpirole versus saline rats [16
]. Moreover, informal examinations of the videotape records were consistent with presence of rituals in quinpirole rats as observed previously [16
]. The final criterion, dependence of checking behavior on environmental context, was not examined explicitly in Phase 1 because it was deemed to be addressed in the Test Phase by virtue of introducing into the open field the unfamiliar and familiar cages.
Home Cage Arrests Locomotion
Because checking behavior involves locomotion to the item(s) of concern, a reduction in the distance of travel provides an indirect index of an attenuation in checking. Figure shows the distance traveled by saline-treated control rats (left column) and by rats treated with quinpirole (right column) during tests when a novel or home-cage was introduced into the open field in the middle of the two-hour session (top row) or immediately at start of testing (bottom row). Inspection of the figure suggests the following. In control rats, both the novel and the home cage produced a small and transient increase in distance traveled when the cage was placed in the open field in the middle of the session; there was little effect when the cages were present from the start of testing. However, in quinpirole rats, the effects produced by the two types of cages were strikingly different. The presence of a home-cage, regardless of the time it was placed into the open field, resulted in a virtual stop of locomotor behavior for about 40 minutes but a novel cage had no effect on distance traveled. Furthermore, although locomotor arrest waned after 40 minutes, the amount of travel did not recover to its usual levels during the test period. This pattern of effects is consistent with the prediction that checking behavior would cease, but only for a limited period of time, in the presence of an option to do something else. The observed differential impact of novel and home cages on locomotor behavior was supported by the statistical analysis below.
Figure 2 The effect of cage familiarity on locomotor behavior in the open field. Arrows indicate the time point at which either a familiar container (the rat's home-cage) or an unfamiliar one (a novel cage) was introduced into the open field. Solid and open circles (more ...)
To simplify statistical analysis, the data were collapsed into two one hour intervals for a 4-way analysis of variance (ANOVA) with 3 repeated measures factors: Cage Familiarity (home-cage vs. novel cage). Time of Cage Introduction (at start of open field test vs. after 60 minutes of open field test) and Hour of Testing (hour 1 of open field test vs. hour 2 of open field test). The 4th factor was the between-subjects factor, Drug Treatment (chronic saline vs. chronic quinpirole). The results of this analysis are shown in Table and confirm the summarized inspection of Figure . Specifically, the differential impact of cage familiarity in quinpirole-treated but not saline-treated rats is suggested by a significant Cage Familiarity by Drug Treatment interaction (p < .001; Table ); inspection of the marginal means (and the associated 95% confidence intervals) showed that in quinpirole rats the distance of travel for the home-cage condition was significantly smaller than for the unfamiliar cage condition (244.3 ± 20.1 m with home-cage present vs. 628.7 ± 26.9 m with novel cage present) and no such difference between these conditions existed in saline controls (46.6 ± 20.1 m vs. 46.3 ± 26.9 m). With respect to the observation that the immediate effect on locomotor behavior was independent of the time at which the home-cage was introduced into the open field, this is supported by similar distances of travel in the immediate hour after placement of the home-cage at either 0 minutes or 60 minutes after start of open field test (46.0 ± 11.5 m vs. 77.3 ± 15.3 m, p > .05)). With regard to the incomplete recovery of locomotor distance, this is supported by the findings of a significant four-way interaction (Table ) and that in the second hour of testing under quinpirole, distance traveled in the presence of the home-cage (placed into the open field at start of the test) was significantly smaller than in the presence of a similarly placed novel cage (229.5 ± 39.5 m vs. 663.9 ± 32.7 m).
Table 1 Summary of statistical analyses for 4 dependent measures. Each dependent variable was analyzed in a Cage Familiarity (CAGE) by Time of Cage Introduction (INTRO) by Hour of Testing (HOUR) by Drug Treatment (DRUG) ANOVA with repeated measures on the first (more ...)
Finally, an additional comparison suggests that locomotor activity under quinpirole increased with the length of exposure to the home-cage. In particular, when the home-cage was introduced into the open field at the 60 minute time point, locomotor distance in hour 2 was 77.3 ± 15.3 m. In contrast, when the home-cage was present from beginning of the test, locomotor distance in hour 2 was significantly more, 229.5 ± 39.5 m. It should be noted that this result is not a drug time-course effect because the two measures are from the same time period, 60-120 min after quinpirole injection.
Staying in Home Cage
In addition to reducing locomotor behavior, the home-cage (but not the novel cage) attracted rats into it. Figure shows the duration of staying in the two types of cages by saline controls and quinpirole-treated rats. As can be seen, both saline and quinpirole rats spent a large portion of the test period inside the home-cage but not the novel cage (1931.4 ± 89.6 s in the home-cage vs. 51.5 ± 14.5 s in the novel cage; for Cage Familiarity, F(1,22) = 419.8, p < .001). However, the duration of staying in the home-cage was significantly longer in the quinpirole rats (2312.5 ± 126.7 s) than in the saline controls (1550.0 ± 126.7 s; for Drug Treatment, F(1,22) = 20.3, p < .001; for Drug Treatment × Cage Familiarity, F(1,22) = 15.1, p = .001).
Figure 3 The effect of cage familiarity on the duration of staying in the cage during a test of open field activity. Arrows indicate the time point at which the rat's home-cage or a novel cage was placed into the open field. Solid and open bars are tests with (more ...)
The 4-way interaction for duration of staying in the cage was also significant (for Cage Familiarity by Time of Cage Introduction by Hour of Testing by Drug Treatment, F(1,22) = 17.0, p < .001; Figure ). This interaction effect seemed related to a between-groups difference. Specifically, during hour 2 (60-120 min intervals in Fig ), quinpirole rats, compared to saline controls, stayed in the home-cage significantly longer in one of the Time of Cage Introduction conditions (cage introduced at 60 min: 3300.6 ± 242.6 s under quinpirole vs. 2158.2 ± 242.6 s under saline, p < .05) but not in the other one (cage introduced at 0 min: 2910.5 ± 258.2 s under quinpirole vs. 2153.1 ± 258.2 s under saline, p > .05).
Resumption of Checking Behavior
As was shown in Figure , in the presence of the home-cage, quinpirole rats begun to move through the open field after a period of arrest. Two kinds of measures indicate that this rise in the distance of travel reflects a resumption of their usual checking behavior. One measure is the frequency of exits from the home-cage (Figure ) and the other one is the frequency of visits to the spot of checking behavior in Phase 1 (Figure ). Together, they suggest that when quinpirole rats began to venture from the home-cage, almost every trip included their previous spot of interest, consistent with performance of checking.
Figure 4 The effect of cage familiarity on the frequency of exits from the cage. Arrows indicate the time point at which the rat's home-cage or a novel cage was placed into the open field. Solid and open bars are tests with the home and the novel cage, respectively. (more ...)
Figure 5 The effect of cage familiarity on the number of visits to the site of the home-base from Phase 1. Arrows indicate the time point at which the rat's home-cage or a novel cage was placed into the open field. Solid and open bars are tests with the home and (more ...)
Exits from cage
Although saline-treated rats spent a large portion of the test period inside the home-cage (Figure ), inspection of Figure (left column) shows that they rarely left it. In contrast, quinpirole rats (Figure , right column) went out from the home-cage and returned to it often, especially in hour 2 (47.4 ± 7.5 exits from home-cage under quinpirole in hour 2 vs. 4.3 ± 7.5 exits from home-cage under saline in hour 2, p < .05; see Table for significant main effect of Drug Treatment, and the triple interaction of Drug Treatment by Cage Familiarity by Hour of Testing).
It is also noteworthy that as for locomotor distance, longer exposure to the home-cage under quinpirole led to a higher number of exits from it (exits from the home-cage in hour 2: cage introduced at 60 min = 66.5 ± 11.7 vs. cage introduced at 0 min = 28.3 ± 4.6, p < .05).
Even though quinpirole rats spent very little time at the location of the novel cage (Figure ), they did come to it sporadically (Figure ), though this number of visits did not reach statistical significance in the comparison between quinpirole and saline rats (11.7 ± 3.3 stops at novel cage under quinpirole vs. 1.7 ± 4.2 stops at novel cage under saline; for Drug Treatment by Cage Familiarity, F(1,22) = 3.84, p = .063).
Visits to previous home base
Figure shows the incidence of visits in each of the 4 conditions to one particular locale in the open field. This locale is the home base from Phase 1 and the then focus of checking behavior. Inspection of the figure shows readily that when a novel cage was introduced into the open field, checking behavior in quinpirole rats still continued to be directed to the same spot. However, as was suggested by the locomotor distance data (Figure ), the incidence of visits to this spot declined precipitously when a novel cage was introduced into the open field. Importantly, when travel through the open field resumed, Figure shows that the number of visits to that spot increased dramatically. Specifically, with the home-cage present from the start of testing, quinpirole rats visited the previous home base nearly 40 times during hour 2. This number was almost as large as the frequency of exits from the home-cage during hour 2 (39.0 ± 11.0 visits to previous home base vs. 66.6 ± 16.6 exits from home-cage, t(11) = 2.17, p = .053), suggesting that quinpirole rats traveled to this place of interest almost every time that they left the home-cage. The special attraction of the former home base is also evident from the observation that the frequency of visits to it was significantly higher than the mean number of stops to any other open field locale (excluding the place containing a cage): 39.0 ± 11.0 visits to previous home base vs. 12.2 ± 8.4 visits per locale, t(11) = 2.8, p = .017.
Finally, as found for other measures, longer exposure to the home-cage under quinpirole led to a higher incidence of checks to the spot of interest (visits to former home base in hour 2: cage introduced at 60 min = 39.0 ± 7.8 vs. cage introduced at 0 min = 13.0 ± 2.9, p < .05).