Betta splendens, better known by the name Betta fish or Siamese fighting fish, belong to the Gourami family and originate from Southeast Asia. They grow to around three inches, and are heavily commercialized due to their eye-catching, vibrant-colored bodies and long, wispy fins (Monvises, et al. 2009). Before their commercialization to western countries, these fish were originally dull in hue and had small fins; this is because they were only bred for fighting at this time (Monvises, et al. 2009). Over time the Betta splendens rose to popularity in small children who would catch them in rice paddies and streams. However, they gained the most popularity for their aggressive, fighting behaviors, such as how males will commonly be seen fighting one another (Curtis 2014). Today both fighting Bettas and display Bettas are bred; the fighting Bettas may live for a few days while the display Bettas can live, on average, for four years (Monvises, et al. 2009).

The Betta fish is a popular fish for observational work because they visually exhibit uniform aggressive acts across the species (Cozza 2016). Aggression is an innate behavior in male Bettas that can be triggered by other males in fights to claim territory, female Bettas for mating, and to protect their eggs (Cozza 2016).

To further understand the link between aggression and male Betta fish, we shifted focus from the Betta to its surrounding environment. When triggered, male Bettas are known to exhibit aggressive motions, such as fin flaring and gill flaring, circling the tank, and charging the tank (Cozza 2016). We will be observing those three distinct behaviors to find a correlation between the amount of aggressive acts and amount of access to a visual stimuli (another male or female betta). We predict that the male Betta splendens will exhibit more aggressive behavior when they have unlimited visual access to other male Bettas, when placed in an empty, controlled tank environment because in a study done at the University of New England, it was found that aggressive behavior was influenced by the audience (being another Betta) and what was in the Betta's tank at the time (Dzieweczynski 2005). Studying aggression in male Betta fish is important because aggression is an almost universal behavior in both human and animal social systems (Filby 2010).

We have also found a link in genetics between the Betta splendens and humans that could potentially link their aggressive behaviors. The HTR1A gene is located on the 5th chromosome in humans, and in both the Betta's and humans, it codes for a receptor for serotonin.Serotonin is a chemical that helps regulate mood (Dayan, et al. 2008). When serotonin levels are depleted in an organism, anxiety, depression, and aggression can be behaviors that the subject organism can experience (Dayan, et al. 2008). We predict that the male Betta splendens and humans will exhibit more aggressive displays of behavior when their serotonin levels are lowered because in an experiment done at University College London, the researchers found that serotonin levels were directly linked to how subjects approached situations. When the levels were lower than normal, organisms reacted to situations presented to them in a more aggressive manor (Dayan, et al. 2008).

Three male Betta splendens and one female Betta splenden were used in the observational study. They were captive bred, and supplied by the LB 144 C-4 lab room. Each fish was maintained singly, about six inches away from any other fish. They were kept in clear, plastic, rectangular containers with a top with holes in it for air circulation throughout their tanks. The dimensions of the tanks were 9in x 5in x 6in. The Betta splendens were fed 4 pellets of Tetra BettaMin fish food once every 24 hours and kept in the C-4 lab room. Before beginning observations it was important to not wear flashy clothing, as this can agitate fish and impact them in a negative way that would reduce the validity of our results.

Before letting two subject fish come into contact with each other, we separated them so they could not see the other fish. This was done by placing large folders to act as blinders in between each of the tanks so as to not disturb them as much as it would have by moving the tanks. This was done because nearly all fishes main sensory systems are based on vision and not touch or sound, as most other land mammals (Rowland). When ready to perform the observation periods, we removed blinders from the desired subject fish and recorded the encounter for five minutes, as visually expressed in Figure 3. We recorded any sudden changes in fin sizes, colors, movements, and positions of the observed fish. Fin sizes can grow extensively during periods of stimulation and aggression (Monvises, et al.).

One observer split both of the two tanks into different halves then recorded where the fish was located throughout the five minute time interval. Another observer was strictly dedicated to the movements of the fish and categorized them into 3 different types of movement. The movements being: gill/fin flaring, charging the tank, and circling the tank. These were the main characteristics observed when performing the first tests. The final observer recorded the 5-minute interaction via video-recording. Directly after this observation period, we immediately set up the blinders in between the tanks and waited for 5 minutes for the fish to readjust. The observers then replicated the specific trial with different combinations of the fish 3 times.

To further test our hypothesis, we paired each male fish with all other male fish and a female fish: Male A was paired with Male B and observed; in the next trial Male A was paired with Male C and observed; in the next trial Male B was paired with Male C and observed. We did the same pairing strategy with the male-to-female observations as well, as seen in Figure 3. The reason we did not only testing male-male behavior but also male-female behavior is because we wanted to prove that the aggressive behavior is dependent on gender. This way our results weren't skewed and our collected data was accurate.

In the second half of our experimental observations, we tested what adding a naturally occurring variable did to the amount of displayed aggression in male betta fish. The pairing strategy stayed the same, however, foliage was added to the tank. The foliage consisted of four inch and two inch wide tall water-plants that came supplied in the C-4 lab room in Holmes Hall. Small sized gravel was also added to the bottom of the tanks. These added variables created an environment for the fish that they would see in their natural environment. After foliage was added, we let the fish have a 10 minute adjusting period so they could get used to the plants and gravel added to their tanks and so they knew the added materials were not a threat. We then performed the same number of trials and different types of pairings as stated previously in the control portion of the observational experiment and recorded our results, observing the same visual aggressive behavior cues as the control portion as well. Each 5-minute trial was recorded digitally via an iPhone 6 and by handwritten tallies on paper, adding a tally each time one of the 3 types of aggressive behavior were displayed. The video recording acted as a playback to make sure the handwritten tally results were accurate and correct between each observer.

As for the human aspect of the observational study, the observers went to a football game at Michigan State University on November 12, 2016. One of the three observers recorded 4, 5-minute intervals of the game. The other two observers used handwritten tallies to observe 3 specific aggressive behaviors in the game of football that include: flagrant personal fouls, false start penalties, and tackling (Jewell, et al.). The video recording acted as a playback to make sure the handwritten tally results were accurate and correct between each observer. All data was checked for significance by performing chi-squared statistical tests.

We predicted that male Betta fish would exhibit more aggressive behavior when they had complete visual access to another male in a controlled environment (Figure 1.1). Over a five minute interval, male Betta fish A and B were video recorded. The video was played back and three different behaviors were tallied over the five minute interval. After twenty seconds passing, male A began to charge the tank and flare its fins. These behaviors were tallied, along with the rest of the aggressive behaviors that occurred in the five minutes. What we can conclude after comparing the tallied behaviors at the end of the five minutes is that male A appeared to be more dominant than male B. While both Bettas did exhibit the four aggressive behaviors at least once, (Figure 1.1) male A exhibited each behavior twice as many times as male B. An environment change did affect the Betta fish (Figure 1.2) but the results were too inconclusive to prove any statistical significance, thus the aggression may have to deal with gender than the environment itself.

In the second portion of the experiment, when we added foliage, we found that there was slightly less aggressive behavior for the most part between the males in the two observational sets because they will have somewhere to hide and avoid conflict. When each male was presented visually with another male, they showed aggressive behavioral cues towards one another for two minutes. After two minutes, the fish were able to distinguish which one was the more dominant one. In each trial, the dominant fish continued to circle his tank and flare his fins in an aggressive manner, claiming his territory; the submissive fish stayed behind and near the foliage for the duration of the experiment (three minutes) and stopped cueing the visual aggressive behaviors.

By performing a test of varied habitats that the fish were not used to, it allowed us to look at the behaviors and how they would change them if their natural habitat was altered. During the first study, the tanks were bare but when we added foliage and gravel to the tanks to create a more “natural” habitat for the fish, the Bettas were found to use their cover extensively. Even if one fish was the more aggressive in the first test between the two, they both used the cover provided as a sort of visual barrier from the opposing fish. When provided with foliage in their tanks, even though most everything in the this part of our study stayed the same, the results varied slightly, we believe. Figure 1.1 and 1.2 show that the statistical bars of the behaviors overlapped on most of the categories that were counted which were slightly discouraging as our predictions hypothesized that they would have a much greater impact on their behaviors.

However, in our female Betta and our male Betta observational set, in Figure 2, the results were more concise and the data showed results that stayed consistent in all of the trials between the three male subjects and the one female. This part of the study was meant to act as a sort of control to see what behaviors were supposed to change when they were to interact with the males. When we changed the environmental factors it helped us figure out what sort of behaviors would change. In Figure 2, we found that the behaviors do change dramatically when the fish are paired with the opposite sex compared to the same sex. The count of behaviors decreased substantially as shown in Figure 2.2.

In our human observation period, the research found that observed behaviors were only observed towards certain players which relates back to the betta fish showing their aggression levels towards only their specific “targets” as in the other males. Those were their direct competitors so it would be wasteful to spend energy trying to dominate irrelevant others.

References

Betta Origins. http://www.bettafishcenter.com/Betta-Origins.shtml. Accessed 16 Oct. 2016.

Cozza, John, and Dana Krempels. Agonistic Behavior in Betta Splendens. 2016. University of Miami, n.d.

Curtis, MacLean, et al. "Siamese Fighting Fish (Betta splendens) Ecological Risk Screening Summary." US Fish and Wildlife Service, Sept. 2014, pp. 3-10. Accessed 16 Oct. 2016.

Dzieweczynski, Earley, et al. "Audience effect is context dependent in Siamese fighting fish, Betta splendens." Behavioral Ecology, vol. 16, no. 6, 29 July. 2005, pp. 1025-1030. Oxford Journals. Accessed 16 Oct. 2016.

Filby, Amy L, et al. "Unravelling the neurophysiological basis of aggression in a fish model." BMC Genomics, 16 Sept. 2010, http://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-11-498. Accessed 17 October 2016.

McGregor, Peter K., Tom M. Peake, and Helene M. Lampe. "Fighting fish Betta splendens extract relative information from apparent interactions: what happens when what you see is not what you get." Animal Behaviour. N.p., Dec. 2001. Web. 15 Oct. 2016. .

Luna, Susan.Betta Splendens Siamese fighting fish. http://www.fishbase.se/summary/Betta-splendens.html#. Accessed 16 October 2016.

Rowland, William J. Studying Visual Cues in Fish Behavior: A Review of Ethological Techniques. Environmental Biology of Fishes, 9 May 1997. Web.

Simpson, M. J. A. "The Display of the Siamese Fighting Fish, Betta splendens." Animal Behavior Monographs. Ed. J. M. Cullen and C. G. Beer. 2013.

Dayan P, et al. (2008) Serotonin, Inhibition, and Negative Mood. PLoS Comput Biol 4(2): e4. doi:10.1371/journal.pcbi.0040004

Figure 1. (1.1, 1.2) Comparisons of aggressive behaviors in male Betta fish depending on their surrounding environment. Figure 1.1 compares the displays of behavior presented between Males A, B, and C when they do not have foliage in their tanks. Figure 1.2 compares the displays of behavior presented between Males A, B, and C when they do have foliage in their tanks. This figure is made to answer the question of how male Betta splendens show aggression towards one another depending on what visual access they have to one another. Their behaviors were observed, categorized, and individually tallied over a five minute time interval. All 5 minute interactions were also video recorded and playbacked. The three aggressive behavior categories observed were gill/fin flaring, circling the tank, and charging the tank. These behaviors were chosen in this study due to previous research into aggressive Betta splenden behaviors done by a group from Mahidol University in Thailand (ScienceAsia). The collected data was averaged, and the standard error was found to show if there was or wasn't statistical significance in the collected data.

Figure 2. (2.1,2.2)Behaviors of male subject Betta fish while in the presence of male Betta fish compared to female Betta fish A. While using the data used in the observational set with the different pairs of male Betta fish in figure 1.1 (shown as 2.1), the results of the same experimental design but instead of two males, one male and one female, are shown in 2.2. The behaviors are the same ones categorized and recorded in the data above. The male Betta fish observed in figure 2.2 were observed, categorized, and individually tallied over a five minute time interval. All 5 minute interactions were also video recorded and played back. The collected data was averaged, and the standard error was found to show if there was or wasn't statistical significance in the collected data between interactions in a male v. male environment and a male v. female environment.

Figure 3.Experimental set-up of the tanks for the observations between the three male(or three female) betta fish. Each observation was done using these pairing strategies for five minutes at a time.These pairing strategies were used to make sure the research team collected data from the most possible sets of different fish combinations. (Each fish was observed interacting with all other fish similarly, regardless of sex.)

Figure 4.