The Effect of Shoal Size on an Individual's Stress Response in Amano Shrimp

Abstract

Shoaling behavior can be frequently observed within Amano shrimp (Caridina multidentata). This experiment was used to assess the intraspecific density-dependent response on an individual's stress response. To test whether the stress response was affected by shoal size, we measured the time spent in a distressed state following isolation and a stress stimulus, and repeated for increased shoal size. Our results demonstrated a strong trend, in that the time spent in a distressed state decreased as the group size increased.

Introduction

Within the literature, there are numerous examples of socially guided behavior throughout a range of non-insect invertebrates (^{Webster and Fiorito, 2001}; ^{Evans et al.}

, 2013). Animals displaying socially guided use information obtained from others within a group/shoal to modify their own behavior. More specifically, social influence (a form of socially guided behavior) is where specific behaviors are either initiated or suppressed due to an interaction with an organism of the same species. This can be triggered by proximity to/observation of a conspecific (^{Webster and Fiorito, 2001}). Resultantly, social influence is likely to be most prevalent within a group dynamic.

Our preliminary experiments noted a degree of social influence within our model organisms Amano Shrimp (Caridina multidentata). It could be observed that the isolation of individuals tended to increase the expression of stressed behavioral patterns (increased rapid, seemingly random movement and a lack of 'settling' behavior). This led to the further observation that organisms within a group showed a faster recovery from a fright stimulus. We, therefore, decided to test whether increasing the shoal size would decrease the recovery time of individual shrimp following a stress stimulus.

The stress stimulus itself would involve time spent in isolation to remove the influence of shoaling/group behavior on an individual's response. It is also assumed that the transfer between tanks would present a sufficient stressful influence.

The hypothesis for this experiment is that there is a negative relationship between the number of shrimp and the time spent in a 'distressed state'. It is predicted specifically that as more shrimp are present in the test tank, the individual will spend less time in a distressed state following the stress stimulus.

Methods

This experiment observed the behavior of Amano Shrimp in a controlled laboratory environment. The general experimental setup is shown below:

Three tanks of identical dimensions were set up at room temperature. Opaque barriers were placed between T1 & T2, and between T2 & T3 to ensure experimental isolation by blocking visual interactions between Shrimp within different tanks. The lack of visual contact between individuals in different tanks was important due to the social influence in non-insect Arthropods being heavily affected by visual stimuli (^{Webster and Fiorito, 2001}).

Tank T1 ('Holding tank') contained 50 individual Amano Shrimp. From the holding tank, one shrimp was transferred to T2 ('Isolation tank') using a small sieve. The individual was then left in isolation for a total of 2 minutes. There was no direct preference in shrimp selection for each trial, as any innate skew, especially towards the sex of the tested shrimp, may have altered the results. Importantly, the literature presented sufficient evidence that there was no significant difference in shoaling behaviors seen between males and females (^{Chapman et al., 2013}).

Following the 2-minute period spent within the Isolation tank, the shrimp was transferred to T3 ('Test tank'). The time taken for the shrimp to recover from the combined isolation/stress stimulus was recorded. This parameter was defined as the time taken for the shrimp to lay in a 'settled' state for a 5-second period (following transferal to the test tank). This procedure was repeated until 10 individuals had been added to the test tank, and the recovery time (i.e., the time spent in the distressed state) was measured only for the shrimp that had undergone the stress stimulus. This gave 10 independent data points for each of the 10 individual shrimp. The experiment was to be repeated 5 times (in its entirety), to give results for 50 different individual shrimp.

To ensure that the results were independent (removing the influence of pseudo-replicates), the shrimp that had previously been tested were moved to a separate tank. Between each repeat of the experiment, the water was changed in the isolation tank and the test tank to ensure any chemical cues/signals released in response to stress/isolation by the previously tested shrimp did not influence the behavior of the shrimp to follow. It was important that chemical communication is controlled within the experiment as scent is the dominant sense used to allow for the influence of social support (^{Webster and Fiorito, 2001}). Failing to remove chemical stimuli from previously stressed individuals may heighten the tested individual's stress response.

Results

The results for the 50 individuals tested were plotted on a scatter graph (each as separate data points). The data for the independent variable (Number of shrimp within the test tank) were plotted on the X-axis, and data for the dependent variable (Time spent in distressed state, seconds) were plotted on the Y-axis. To allow for linear regression analysis, the graph has a regression line plotted.

A simple linear regression analysis was calculated to predict that 'the time spent in distressed state' was based on 'the number of Shrimp within the test tank'. A significant regression equation was found: (F(1,48) = 35.23, p < 0.0001), with an R² of 0.423. Individuals' predicted time spent in the distressed state is equal to (25.620) + (-2.044) (number of Shrimp within the test tank) seconds. This means that the time spent in the distressed state decreased by -2.044 for every individual shrimp added.

Discussion

The hypothesis of the lab was that "there is a negative relationship between the number of shrimp and the time spent in a distressed state". The data supported this hypothesis since our analysis presented a strong, significant negative linear trend (regression coefficient b < 0). It can be concluded that, within our experiment, the Amano shrimp on average displayed less distressed behavior with an increased number of shrimp present in the test tank. As the p-value is less than the significance level of 0.05, the data provide sufficient evidence to conclude that the linear regression model shows a significant association between the two variables (time spent in distressed state & number of shrimp in the test tank).

Social influence appeared to suppress stressed behavioral patterns, showing some level of density dependence. As the results showed some variance, it could be suggested that social influence is stronger in some individuals. This may be due to different behavior types/personalities.

Many of our own conclusions and those made within the literature assume that the shoaling behavior observed is ultimately to reduce predation. It may be difficult to accurately deduce whether the behavior observed within the experiment is a defense strategy against predators (such as the dilution effect) due to a number of confounding potential advantages of this grouping behavior. It is suggested that shoaling behaviors in some species present advantages within food gathering and reproduction (^{Foster & Treherne, 1981}). This, however, would not fully explain the reduction in the stress response by the shrimp. If the behavior is concluded to be a predatory defense strategy, the mechanism behind this may not be as simple as an individual's increase in fitness due to the dilution effect alone. There are a number of ways that shoaling mechanisms can defend an individual from predators, whether this is through the 'confusion effect' (^{Landeau & Terbough, 1986}) or simply allowing greater success in identification of a potential predator (^{Foster & Treherne, 1981}). As all of the listed advantages of shoaling behavior in predator defense could aid in explaining the ultimate cause of the observed trend, it may also be informative to conduct additional experiments assess the exact mechanism underlying the behavior.

References

1. Chapman, B., Hegg, A. and Ljungberg, P. (2013). Sex and the Syndrome: Individual and Population Consistency in Behavior in Rock Pool Prawn Palaemon elegans. PLoS ONE, 8(3), p.e59437.
2. Evans, S R, Finnie M and Manica A (2007) Shoaling preferences in decapod crustacean. Animal Behaviour 74: 1691-1696.
3. Foster, W. and Treherne, J. (1981). Evidence for the dilution effect in the selfish herd from fish predation on a marine insect. Nature, 293(5832), pp.466-467.
4. Landeau, L. and Terborgh, J. (1986). Oddity and the 'confusion effect' in predation. Animal Behaviour, 34(5), pp.1372-1380.
5. Webster, S. and Fiorito, G. (2001). Socially guided behavior in non-insect invertebrates. Animal Cognition, 4(2), pp.69-79.