The Psychological Record, 1997,47, 167-174 EFFECT OF ILLUMINATION CONDITION ON RISK ASSESSMENT BEHAVIORS OF MICE ERNEST D. KEMBLE and MICHAEL J. GOBLIRSCH University of Minnesota-Morris Levels of risk assessment evoked by either a synthetic predator odor or noncontact exposure to live rats were examined under white or red light conditions. Defensive responses to the predator odor were infrequent in white light and showed a modest, but reliable, increase under red light. Noncontact exposure to rats, in contrast, produced considerably higher levels of risk assessment than did odors in white light, and these levels were nearly doubled under red light. Judicious selection of eliciting and background stimuli may provide considerable control over the intensity of risk assessment. Such procedures may prove useful for the study of anxiety-like states. Risk assessment is a popular tool for the study of anxiety-like states (e.g., Blanchard & Blanchard, 1988, 1990) but the variables controlling the intensity of these behaviors are not well characterized. Presentation of various odors, for example, regularly evokes modest elevations in the levels of these defensive responses (e.g., Blanchard et ai., 1990; Garbe, Kemble, & Rawleigh, 1994; Williams & Scott, 1989) which are sensitive to both anxiolytic and anxiogenic drug treatments (Blanchard et ai., 1990; Blanchard, Tauklis, Rodgers, Magee, & Blanchard, 1993; Garbe, Kemble, & Strunk, 1993; Kemble & Gordon, 1996; Zangrossi & File, 1992). Although this experimental manipulation is clearly useful, it should be noted that a reasonably wide range of both predator and nonpredator odors have roughly equivalent effects on risk assessment (Garbe et ai., 1994). Furthermore, repeated prior defeat experience (Kemble & Bolwahnn, in press) does not increase odor reactivity. If odorinduced risk assessment is generally insensitive to major experimental manipulations, then its usefulness for the study of defensiveness may be somewhat compromised. Alternatively, it is possible that some relevant stimulus conditions remain untested. Illumination condition appears to be a likely candidate. Not only is open field activity affected by illumination levels (e.g., Welker, 1959) but Lester and Faneslow (1992) find that light We thank Pherotech Inc. for donating the synthetic predator odorant which was used in Experiment 1. Reprint requests may be sent to Ernest Kemble, Division of Social Sciences, University of Minnesota-Morris, Morris, MN 56267.
168 KEMBLE AND GOBLIRSCH and dark conditions differ in their ease of association with electric shock. Because our previous experiments were invariably conducted under white light illumination, it seemed possible that this lighting condition suppressed risk assessment to some extent. If so, substitution of red for white light might elevate levels of defensiveness further. The following experiments explored this possibility. General Method Subjects The subjects were experimentally naive male CD-1 mice weighing 36.7-54.7 g. The subjects were individually housed in 29.5- x 18.5- x 12.0-cm clear polycarbonate cages having a wood shavings substrate. The mice were housed under a 12-hr light/dark cycle and tested during the light phase. All subjects received ad lib access to Purina Lab Chow and water throughout the experiment. The mice were randomly assigned to weight balanced groups designated for testing under white or red fluorescent illumination. Apparatus Testing was conducted in a 46.0- x 24.0- x 21.0-cm clear polycarbonate cage covered with a clear Plexiglas sheet. The cage was divided into 29.5- x 24.0- x 21.0-cm and 16.5- x 24.0- x 21.0-cm compartments by a 0.6 cm wire mesh screen. The fear-evoking stimuli were placed in the larger compartment and the mice in the smaller. Illumination was provided by 6 white or 6 red 34-W fluorescent tubes mounted 1.7 m above the testing compartment. The frequency and duration of risk assessment behaviors were scored with the aid of a computer data acquisition system. Procedure The mouse and fear-evoking stimulus were placed in their respective compartments under the designated illumination condition. The frequency and duration of four risk assessment behaviors (defensive burying, immobility, stretched attention, and flatback approach) as previously described by Blanchard et al. (1990) were then recorded for 10 min by an observer quietly seated 1.0-1.2 m from the test chambers. We found that levels of risk assessment recorded in this way were virtually identical (all ps >.10) to those obtained when behaviors were observed through a one-way vision screen. When possible, groups were compared by means of t tests. Some behaviors were absent among substantial numbers of subjects, however, and the groups were compared by Mann-Whitney U tests. Experiment 1 As mentioned previously, a variety of predator and nonpredator odors produce similar moderate elevations of risk assessment when testing is
ILLUMINATION CONDITION AND RISK ASSESSMENT 169 8 7 (.) 6-5 L. w. 4 (C 3 I: 2 1 p<.10 p<.oo5 (.) '" "-' 5 4.-.., (C 3 L. 2 (C I: 1 White Red LIGHT [ONDITION Figure 1. Mean frequency (upper panel) and duration (lower panel) of stretched attention under white or red light conditions. conducted under white light (Garbe et ai., 1994). In this experiment risk assessment was measured under either white or red fluorescent illumination in response to a synthetic predator odorant. This stimulus effectively deters feeding for extended periods in field settings (Sullivan, Crump, & Sullivan, 1988) and produces slight but significant elevations of risk assessment in white light (Kemble & Bolwahnn, in press).
170 KEMBLE AND GOBLIRSCH Method Eighteen mice were tested under either white (N 9) or red (N 9) light conditions. Synthetic predator odor was provided by a 5.0- x 0.3-cm slow release cylinder containing a 1: 1 mixture of 2-propylthietane (PT) and 1, 2 dithiolane (PDT). This pungent odorant reduces plant feeding by two species of meadow voles in the field (Sullivan et ai., 1988). The odor source was placed in a glass petri dish in the center of the larger compartment 30 min before testing. Results The mean frequency of stretched attention is depicted in the upper panel of Figure 1. Testing under red light significantly increased the frequency of this behavior, U(9, 9) 6, P <.005. Durations (lower panel) were also increased by red light but group differences attained only a marginal level of significance (p < 10). Although there was also a slight increase in the frequency (White, M 1.3; Red, M 2.2) and duration (White, M 0.7 sec; Red, M 1.6 sec) of defensive burying, these differences were not statistically reliable (ps >.10). The remaining risk assessment behaviors were quite infrequent in both groups with no suggestion of reliable group differences (ps>.10). Experiment 2 As Experiment 1 makes clear, testing under red light does indeed increase odor-induced risk assessment. This procedure might therefore be useful for experimental manipulations in which slightly higher baseline levels of defensiveness are desirable. As in previous experiments, however, levels of defensiveness to the odorant were rather low, even under red light, and the effects were restricted to a single risk assessment behavior. In contrast, exposure of rats to live cats evokes intense and protracted risk assessment (e.g., Adamec & Shallow, 1993; Blanchard & Blanchard, 1988). Although this procedure clearly produces a major elevation of defensiveness, the great intensity and duration of the behaviors may also reduce its usefulness for some experimental applications (e.g., some drug treatments). A procedure which induces more moderate effects would, therefore, seem desirable. Preliminary observations suggested that live rats in white light produce substantial, but short-term, elevations in risk assessment. This experiment therefore explored the possibility that rats produce higher levels of defensiveness than odors, and that such elevations are modulated by illumination condition. Method Subjects The subjects were 18 mice which were tested under white (N 9) or
ILLUMINATION CONDITION AND RISK ASSESSMENT 171 50 (.) 40 - L. 30 ra 20 10 p<.005 p<.oo 1 (.) 50 40.-.., 30 L. 20 10 Red LIGHT CONDITION White Figure 2. Mean (SEM) frequency (upper panel) and duration (lower panel) of stretched attention under white or red light conditions. red (N 9) light. Three male and 2 female hooded rats weighing 410-530 g served as fear-evoking stimuli. Apparatus and Procedures The apparatus was that used in Experiment 1. At the beginning of the test, one rat and one mouse were simultaneously placed in the
172 KEMBLE AND GOBLIRSCH apparatus. The wire mesh barrier prevented contact between mouse and rat. Each rat served as the fear-evoking stimulus for 3 or 4 mice. Otherwise, procedures were also identical. Results The frequency of stretched attention is depicted in the upper panel of Figure 2. It can be seen that rats evoked substantially higher levels of this behavior in white light than did the predator odor, and that testing under red light roughly doubled its frequency, t(16) 7.56, P <.001. Stretched attention durations showed a similar increase under red light, t(16) 3.75, P <.005. The frequency, but not duration (p >.10), of immobility (White, M 6.4; Red, M 11.2) was also increased by red light, 1(16) 2.16, P < 05. Although the frequency (White, M 11.8; Red, M 18.3) and duration (White, M 12.4 sec; Red, M 22.6 sec) of flatback approaches were somewhat elevated under red light, these differences were not statistically reliable (ps >.10). Both the frequency and duration of defensive burying were similar in the two groups with no suggestion of reliable group differences (ps >.10). Discussion Risk assessment, evoked by either odors or a live rat, was clearly potentiated when red light was substituted for white. Because the mice exposed to red light in these experiments were tested during the 12-hr light portion of the light/dark cycle, it is not clear whether the effects were due to red light per se or to the abrupt transition from white to red illumination. Testing during the normally dark portion of the light/dark cycle would be useful in addressing this issue. In any case, this manipulation does clearly elevate responsiveness and should increase the range of experimental applications which can be fruitfully investigated with this paradigm. Although it is conceivable that a higher concentration of the predator odor might produce,similar effects, we (Garbe et ai., 1994) found that four novel odorants, differing considerably in pungency, produced similar modest elevations of defensiveness. Moreover, additional (unpublished) research by the above authors failed to reveal any further elevations when the amount of one odorant (citronella) was increased by a factor of 10. Taken together, the data suggest that a reasonably wide range of odors evoke only mild elevations of risk assessment behaviors. Because live rats provide a rich array of stimuli, those responsible for elevated defensiveness obviously cannot be identified from these data. Further research employing more narrowly defined partial predator stimuli should be useful for this assessment. Evans, Macedonia, and Marler (1993), for example, find that predator-like visual stimuli evoke both vocal and nonvocal fear responses in chicks whose levels are dependent on both the speed and apparent size of the stimulus.
ILLUMINATION CONDITION AND RISK ASSESSMENT 173 Moving/looming stimuli also increase defensiveness of rats (Blanchard, Mast, & Blanchard, 1975) and a considerable range of other species (Ydenberg & Dill, 1986). In addition, the calls of predatory birds increase defensiveness in both mice (Hendrie & Neill, 1991) and birds (Hauser & Caffrey, 1994). Systematic investigation of such stimulus variables may be useful in identifying the most effective fear-inducing stimuli and further contribute to our understanding of anxiety-like states in animals. This possibility is currently under investigation. References ADAMEC, R. E., & SHALLOW, T. (1993). Lasting effects on rodent anxiety of a single exposure to a cat. Physiology and Behavior, 54, 101-109. BLANCHARD, D. C., & BLANCHARD, R. J. (1988). Ethoexperimental approaches to the biology of emotion. In M. R. Rosenzweig & L. W. Porter (Eds.), Annual Review of Psychology, 39, 43-68. BLANCHARD, R. J., & BLANCHARD, D. C. (1990). An ethoexperimental analysis of defense, fear, and anxiety. In M. McNaughton & G. Andrews (Eds.), Anxiety (pp. 124-133). University of Otago Press: Dunedin. BLANCHARD, R. J., BLANCHARD, D. C., WEISS, S. M., & MEYER, S. (1990). Effects of ethanol and diazepam on reactivity to predatory odors. Pharmacology, Biochemistry and Behavior, 35, 775-780. BLANCHARD, R. J., MAST, M., & BLANCHARD, D. C. (1975). Stimulus control of defensive reactions in the albino rat. Journal of Comparative and Physiological Psychology, 88, 81-88. BLANCHARD, R. J., TAUKLlS, H. R., RODGERS, R. J., MAGEE, L. K., & BLANCHARD, D. C. (1993). Yohimbine potentiates active defense responses to threatening stimuli in Swiss-Webster mice. Pharmacology, Biochemistry and Behavior, 44, 673-681. EVANS, C. S., MACEDONIA, J. M., & MARLER, P. (1993). Effects of apparent size and speed on the response of chickens, Gallus gallus, to computergenerated simulations of aerial predators. Animal Behavior, 46, 1-11. GARBE, C. M., KEMBLE, E. D., & RAWLEIGH, J. M. (1994). Novel odors evoke risk assessment and suppress appetitive behaviors in mice. Aggressive Behavior, 19,447-454. GARBE, C. M., KEMBLE, E. D., & STRUNK, P. C. (1993). Effects of chlordiazepoxide on odor-induced risk assessment behaviors in mice. Bulletin of the Psychonomic Society, 31, 314-316. HAUSER, M. D., & CAFFREY, C. (1994). Anti-predator response to raptor calls in wild crows, Corvus brachrhynchos. Animal Behaviour, 48,1469-1471. HENDRIE, C. A., & NEILL, J. C. (1991). Exposure to the calls of predators of mice activates defensive mechanisms and inhibits consummatory behaviour in an inbred mouse strain. Neuroscience and Biobehavioral Reviews, 15, 479-482. KEMBLE, E. D., & BOLWAHNN, B. L. (in press). Immediate and long term effects of novel odors on risk assessment in mice. Physiology and Behavior.
174 KEMBLE AND GOBLIRSCH KEMBLE, E. D., & GORDON, C. J. (1996). Yohimbine increases novel odorinduced risk assessment behaviors. The Psychological Record, 46, 179-186. LESTER, L. S., & FANESLOW, M. S. (1992). Nocturnality as a defensive behavior in the rat: An analysis in terms of selective association between light and aversive stimulation. The Psychological Record, 42, 221-253. SULLIVAN, T. S., CRUMP, D. R., & SULLIVAN, D. S. (1988). Use of predator odors as repellents to reduce feeding damage by herbivores. III. Montane and meadow voles (Microtus montanus and Microtus pennsylvanicus). Journal of Chemical Ecology, 14,363-377. WELKER, W. I. (1959). Escape, exploratory and food seeking responses of rats in a novel situation. Journal of Comparative and Physiological Psychology, 52, 1 06-111. WILLIAMS, J. L., & scon, D. K. (1989). Influence of conspecific and predatory stressors and their associated odors on defensive burying and freezing responses. Animal Learning and Behavior, 17, 383-393. YDENBERG, R. G., & DILL, L. M. (1986). The economics of fleeing from predators. In J. S. Rosenblatt, C. Beer, M. Busnel, & P. J. B. Slater (Eds.), Advances in the Study of Behavior (vol. 37, pp. 305-317). New York: Academic Press. ZANGROSSI, H., & FILE, S. E. (1992). Chlordiazepoxide reduces the generalized anxiety, but not the direct responses, of rats exposed to cat odor. Pharmacology Biochemistry and Behavior, 43, 1195-1200.