Printer Friendly

Long-term consequences of agonistic interactions between socially naive juvenile American lobsters (Homarus americanus).

Though aggression has been studied extensively, little is known about its underlying neural mechanisms. The purpose of the present study was to examine the persistence of behavioral effects of fighting experience in socially naive juvenile lobsters (Homarus americanus Milne-Edwards, 1837), as a step toward investigating both the neural substrate and ethological relevance of such effects. In socially naive juvenile lobsters, fighting behavior is highly stereotypical, easily evoked, and readily quantifiable (1). In the wild, fights are often short, as size asymmetries and previous experience play an important role in early decision-making (2, 3). In the laboratory, stable dominance relationships form readily between pairs of lobsters (1, 2, 4), and hierarchies form within groups (5, 6, 7, 8).

Previous studies have shown that subordinates tend to avoid engaging dominants (1, 3, 5, 6, 9), and if interactions do take place, their duration and intensity are reduced. Thus, with the formation of dominance relationships, fighting tapers off, and the duration and intensity of encounters can be used as measures of "willingness" to fight (1). Studies of adult, socially experienced lobsters indicate that recognition of a dominance relationship is retained for at least 1 day of separation after an initial fight (4). Pairs separated for 7 days showed variable results, and all animals seemed to have forgotten their initial dominance relationships after 14 days. Although the patterns of fighting behavior between pairs of experienced and pairs of socially naive animals are similar, differences are apparent, for example, in the time taken to establish dominance and in the intensity of fighting. Karavanich and Atema (4) postulate that these differences may be due to shaping of the adult behavior by experience. Naive lobsters exhibit more ritualized fighting, making it possible to quantify the behavior and to define the consequences of experience on the subsequent patterns of behavior (1).

We have characterized the conduct of pairs of socially naive, juvenile animals over three 30-min fighting periods. The first fight served to establish the dominance relationship, the second (1 h later) was used to confirm that relationship, and the third (either 1, 4, or 7 days later) was used to determine how long the effects of the initial two fighting periods continued to influence subsequent fighting behavior. A preliminary report of these data has been published (10).

Juvenile lobsters, that had been isolated visually and physically, but not chemically, from the fourth post-larval stage were obtained from a rearing facility at the New England Aquarium, Boston, Massachusetts. Animals were weighed to determine their size (1.04-3.49 g) and were molt-staged based on exoskeletal flexibility. Only intermolt animals were used because agonistic behavior varies over the molt cycle (11, 12). After transport to a 15-16 [degrees]C room with a 12-h light/dark cycle at Harvard Medical School, lobsters were again visually and physically but not chemically isolated, and housed in perforated opaque containers in circulating artificial seawater (Marine Biotech, salinity 33 ppt). They were fed squid, shrimp, or fish three times a week. Animals were isolated in individual 9-1 tanks the day before their first fights.

The following three experimental groups of lobsters were assembled randomly: animals separated for 1 day (7 pairs), animals separated for 4 days (7 pairs), and animals separated for 7 days (6 pairs). Within-pair differences in weight ranged from 9% to 29%. A one-way parametric analysis of variance (ANOVA) confirmed that the mean percent weight difference between individuals in a pair did not vary significantly between the three experimental groups (P > 0.05). Under these conditions, animals usually establish a hierarchical relationship during their first 30-min fighting period (1). Pairs were selected at random, paying no attention to gender because in animals of the early developmental stage used, claws are relatively undifferentiated between the sexes (8, 13, 14). Difference in claw size is a major factor conferring fighting advantage in adult male lobsters (6). Formation of a dominance relationship was determined based on consistent retreats in the first fight by one animal, and only pairs that exhibited a stable dominance relationship between the first and second fight were paired for a third fight.

The fights were set up in the following manner: two animals were removed from their isolation tanks; placed on either side of an opaque divider in a 38-1 observation tank with a gravel-lined floor; and allowed to acclimate, with a charcoal filter running, for 30 min. The divider was removed for 30 min (Fight 1), replaced for 1 h, and removed again for 30 min (Fight 2), after which the opponents were returned to their isolation tanks. Pairs were separated for 1, 4, or 7 days, after which the animals were acclimated in the fight tank for 30 min, and then allowed to fight for 30 min (Fight 3). The filter in the observation tank was turned off whenever the divider was removed. Artificial seawater was changed and filtered between each pair, and all fights were videotaped using a Panasonic 3260 or PV-A208 video camera. Only physically intact animals were used, and these were not re-used after a series of three fights.

During fight periods, opponents came together, interacted, and then separated. Each such event was called an "encounter," and these occurred repeatedly during a fight. Each encounter was scored from the time two animals came within one body length of each other to when one or both broke off and they remained separated for at least 5 s. Measurements taken included encounter duration and maximum intensity level reached during the encounter (on a scale from 0 to 3; see (1)), which animal approached, and which animal retreated. Opponents were distinguished by their exoskeletal markings and coloring. Data were entered into FileMaker Pro and analyzed using Microsoft Excel or OriginLab 7.0.

We analyzed changes in the mean duration of individual encounters within 30-min fighting periods, as well as the summed duration of all encounters within 30-min fighting periods. For the analysis of fight intensity, we divided the intensity scale into two bins: level 0, when only one animal is engaged in offensive aggressive behavior; and levels 1-3, when both animals are engaged in offensive aggressive behavior. The percentage of encounters that reached a maximum intensity greater than 0 was then compared across fighting periods. Nonparametric Friedman ANOVAs were used to detect significant differences in encounter duration and intensity across the three fighting periods for each group, followed by post hoc Wilcoxon signed-rank tests for all three possible pairwise comparisons of fights within a group.

Figure 1 shows the durations of all encounters for the three groups of animals over all fights. The greatest variation in the duration of individual encounters occurs in about the first 15 encounters of the initial fight within each group (Fig. 1, left). As the initial fight progressed and dominance was established, the duration of individual encounters was reduced. To test whether fighting behavior changed significantly after different periods of separation, we used Friedman ANOVAs for within-group differences across Fights 1, 2, and 3 in the three groups of animals. The data for mean duration, summed encounter duration, and percent of encounters with maximum intensity greater than 0 are presented in Table 1 and are represented graphically in Figure 2.

The mean duration of encounters was significantly lower in the third fight when compared to the initial fight for animals separated for 1 day (ANOVA: P < 0.025; post hoc Fight 1 vs. 3: P < 0.05) and 4 days (ANOVA: P < 0.025; post hoc Fight 1 vs. 3: P < 0.05), but not 7 days (ANOVA: P < 0.01; post hoc Fight 1 vs. 3: 0.05 < P < 0.10). The same trend is demonstrated with the summed duration data (Table 1). Thus, animals separated for 1 or 4 days fought less during their third fight than they did during their first fight. Apparent differences in encounter durations during Fight 1 between groups (i) could not be compared directly by Friedman ANOVA due to the different number of pairs fought in the group separated for 7 days (n = 6) and the other two groups (n = 7); and (ii) likely result from chance, as initial fights for all three experimental groups were conducted in parallel and animals were assigned to groups at random. The maximum intensity data yielded a consistent finding for pairs separated for 1 day (ANOVA: P < 0.05; post hoc Fight 1 vs. 3: P < 0.05); however, there is no significant difference for pairs separated for 4 or 7 days. No significant differences were found in the post hoc comparisons of Fight 2 to either Fight 1 or Fight 3 for any measurements within any group.


In many species, agonistic experience has long-lasting effects on the subsequent behavior of both winners and losers (15). Here, we focus on the persistence of such effects in pairs of socially naive juvenile lobsters allowed to interact under controlled conditions. Our results show that two short, 30-min fights can exert significant long-lasting effects on the fighting behavior of socially naive juvenile lobsters. Due to their extended duration, we refer to them as "memory" effects, although we have not addressed their underlying mechanisms--for example, whether these effects indicate a memory of hierarchical decisions made or are specific to the individual familiar opponent remains an open question.

Our results using juvenile animals have similarities to those reported in studies of adult socially experienced lobsters (4). In the latter studies, animals fought with decreased agonistic intensity against familiar opponents but not unfamiliar opponents 1 day after their initial fights. After 7 days, results were variable and differences were not statistically significant; after 14 days there was no evidence of any memory of status. Here, we have studied dominance recognition in socially naive juvenile lobsters. These animals differ not only in age but also in social experience from wild-caught adult lobsters. In our experiments, they fought with their first and only opponent. This is a critical distinction, as the previous experience of adult lobsters likely influences their behavior in subsequent interactions. Our study did not address the issue of individual recognition.


Although a wide time range has been observed in the persistence of memory of social status in other species (15), two generalizations have emerged. The first is that memory of status lasts longer in losers of fights than in winners. This result has been suggested to indicate the greater importance of an animal knowing that it is a poor fighter for surviving future encounters in the wild. The second generalization is that animals that have fewer encounters in the wild remember the outcome of those encounters for longer periods of time. That is, to apply its past experiences adaptively, an animal with few encounters requires a longer memory.

Our study and others (4) indicate that lobsters retain memory of social status for a relatively long time. Field observations show that adult lobsters spend much of their lives isolated in shelters, encountering conspecifics largely when foraging for food (16, 17). Though these studies targeted specific sub-populations of lobsters, the relatively low frequency of agonistic encounters observed is consistent with the idea that lobsters should retain a relatively long memory of status. The agonistic behavior of juvenile lobsters in the field has yet to be studied. The duration of memory retention found in this study suggests that changes in gene expression might underlie these changes in behavior, since it is unlikely that purely second-messenger-mediated mechanisms would last for days. Our present results, however, yield no insights into the nature of the biological changes that might take place. What they do offer is information on the time window during which such changes should be sought.
Table 1 Comparison of Fight 1, Fight 2, and Fight 3

Days of separation  Fight 1                    Fight 2

Mean Encounter Duration (seconds)
      1              47.09 ([+ or -] 56.35)     13.74 ([+ or -](9.67)
      4              35.30 ([+ or -] 19.54)     10.53 ([+ or -] 7.70)
      7              22.11 ([+ or -] 13.68)     10.85 ([+ or -] 9.46)

Summed Encounter Duration (seconds)
      1             579.71 ([+ or -] 177.812)  247.57 ([+ or -] 70.06)
      4             811.29 ([+ or -] 195.95)   219.71 ([+ or -] 55.80)
      7             443.83 ([+ or -] 120.28)   189.33 ([+ or -] 90.09)

Percent of Encounters with Maximum Intensity > 0
      1              77.51 ([+ or -] 10.44)     56.32 ([+ or -] 25.75)
      4              64.92 ([+ or -] 32.58)     54.03 ([+ or -] 27.37)
      7              66.58 ([+ or -] 20.87)     70.40 ([+ or -] 16.11)

                                              Friedman   signed-rank
Days of separation  Fight 3                   ANOVA      (Fight 1 vs. 3)

Mean Encounter Duration (seconds)
      1               5.92 ([+ or -] 2.63)    P < 0.025     P < 0.05
      4               8.57 ([+ or -] 8.12)    P < 0.025     P < 0.05
      7               7.16 ([+ or -] 7.58)    P < 0.01   0.05 < P < 0.10

Summed Encounter Duration (seconds)
      1             124.14 ([+ or -] 36.19)   P < 0.05      P < 0.05
      4             204.00 ([+ or -] 107.19)  P < 0.005     P < 0.05
      7             111.83 ([+ or -] 59.30)   P < 0.01   0.05 < P < 0.10

Percent of Encounters with Maximum Intensity > 0
      1              38.90 ([+ or -] 23.02)   P < 0.05      P < 0.05
      4              55.04 ([+ or -] 30.47)     NSD           N/A
      7              45.90 ([+ or -] 34.29)     NSD           N/A

Mean encounter duration, summed encounter duration, and percent
encounters with a maximum intensity > 0 ([+ or -] standard deviation)
for Fight 1 (initial), Fight 2 (1 hour later), and Fight 3 (1, 4, or 7
days later). NSD = No significant difference; N/A = not examined because
Friedman ANOVA yielded no significant difference.


We thank Mr. Jason Goldstein and Dr. Marianne Farrington for supplying us with juvenile animals from a rearing facility at the New England Aquarium, Dr. Margaret Bradley for advice and assistance pertaining to experimental setup and animal care in our laboratory at Harvard Medical School, and Dr. Mayumi Morimoto for statistical consultation. This research was supported by a grant from the National Science Foundation (IBN-0090730 to EAK).

Received 9 July 2004; accepted 25 August 2004.

Literature Cited

1. Huber, R., and E. A. Kravitz. 1995. A quantitative analysis of agonistic behavior in juvenile American lobsters (Homarus americanus). Brain Behav. Evol. 46: 72-83.

2. Scrivener, J. C. 1971. Agonistic behavior of the American lobster, Homarus americanus. Fish. Res. Board. Can, Tech. Rep. 235: 1-113.

3. Atema, J., and J. S. Cobb. 1980. Social behavior. Pp. 409-450 in The Biology and Management of Lobsters, Vol. 1, B. F. Philips, ed. Academic Press, New York.

4. Karavanich, C., and J. Atema. 1998. Individual recognition and memory in lobster dominance. Anim. Behav. 56: 1553-1560.

5. Stein, L., S. Jacobson, and J. Atema. 1975. Behavior of lobsters (Homarus americanus) in a semi-natural environment at ambient temperatures and under thermal stress. Woods Hole Oceanogr. Inst. Tech. Rep. 75.

6. Jacobson, S. M. 1977. Agonistic behavior, dominance and territoriality in the American lobster, Homarus americanus. Ph.D. dissertation, Boston University, Boston, MA.

7. Sastry, A. N., and R. E. Ehinger. 1980. Dominance hierarchies among communally held juvenile lobsters, Homarus americanus. Mar. Behav. Physiol. 7: 85-93.

8. Atema, J., and R. Voight. 1995. Behavior and sensory biology. Pp. 331-348 in Biology of the Lobster Homarus americanus, J. R. Factor, ed. Academic Press, San Diego.

9. Karavanich, C., and J. Atema. 1998. Olfactory recognition of urine signals in dominance fights between male lobster, Homarus americanus. Behaviour 135: 719-730.

10. Rutishauser, R. L., E. J. Wilkinson, A. E. Hower, A. Delago, S. I. Cromarty, R. Huber, B. S. Beltz, and E. A. Kravitz. 1999. Agonistic behavior in lobsters: persistence of fight-induced changes in status and modulation by serotonin. Abstract Program No. 32.33. Society for Neuroscience, Miami, FL.

11. Tamm, G. R., and J. S. Cobb. 1978. Behavior and crustacean molt cycle--changes in aggression of Homarus americanus. Science 200: 79-81.

12. Cromarty, S. I., J. S. Cobb, and G. Kass-Simon. 1991. Behavioral analysis of the escape response in the juvenile lobster Homarus americanus over the molt cycle. J. Exp. Biol. 158: 565-581.

13. Lang, F., C. K. Govind, W. J. Costello, and S. I. Greene. 1977. Developmental neuroethology--changes in escape and defensive behavior during growth of lobster. Science 197: 682-685.

14. Govind, C. K. 1992. Claw asymmetry in lobsters: case study in developmental neuroethology. J. Neurobiol. 23: 1423-1445.

15. Hsu, Y., and L. L. Wolf. 1999. The winner and loser effect: integrating multiple experiences. Anim. Behav. 57: 903-910.

16. Karnofsky, E. B., J. Atema, and R. H. Elgin. 1989. Field observations of social behavior, shelter use, and foraging in the lobster, Homarus americanus. Biol. Bull. 176: 239-246.

17. Karnofsky, E. B., J. Atema, and R. H. Elgin. 1989. Natural dynamics of population structure and habitat use of the lobster, Homarus americanus, in a shallow cove. Biol. Bull. 176: 247-256.


(1) Harvard Medical School, Department of Neurobiology, 220 Longwood Avenue, Boston, Massachusetts 02115; and (2) Assumption College, Department of Natural Sciences, 500 Salisbury Street, Worcester, Massachusetts 01609-1296

* To whom correspondence should be addressed: E-mail:
COPYRIGHT 2004 University of Chicago Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2004 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rutishauser, Rachel L.; Basu, Alo C.; Cromarty, Stuart I.; Kravitz, Edward A.
Publication:The Biological Bulletin
Date:Dec 1, 2004
Previous Article:Composition of a one-year-old Riftia pachyptila community following a clearance experiment: insight to succession patterns at deep-sea hydrothermal...
Next Article:Histidine suppresses zinc modulation of connexin hemichannels.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters