The Effect of Environmental Enrichment on the Foraging Behavior of

Garter Snakes - Thamnophis sirtalis


Ramona Puskar




Enrichment may affect the foraging behavior of Garter Snakes (Thamnophis sirtalis). In this study, I investigated the effects of environmentally enriched and standard laboratory housing on foraging behavior in garter snakes (Thamnophis sirtalis). The snakes were put into a test aquarium once a week and timed to see how long it took them to find their prey in this new enriched environment. There were no significant differences between the enriched snakes and the control snakes revealed through this task. Therefore this study demonstrated that the differences in housing for garter snakes do not affect their foraging behavior.



Environmental enrichment is the act of making the environment where an organism is living, playing or working in better. This organism can be a tree dwelling koala bear, an adult human, an eight week old puppy or even a snake. Their environment can be a eucalyptus tree in a zoo, an office cubicle, a house with a fenced in back yard or even a glass cage with an escape proof lid. Their enrichment can include a very leafy tree, a calendar with beautiful beach settings, a snack of peanut butter in a hollowed out bone, or a hollowed out half of a coconut and some rocks. It doesn’t matter what as long as the initial environment is made better in some way. This study looked at the effect of environmental enrichment on the foraging behavior of garter snakes.

            Foraging behavior is the act of searching for food. A koala will look in eucalyptus tree, a human will search in a refrigerator, a puppy will look at its master with woeful eyes and a snake will use its vomeronasal organ to sense the chemical in the air which will lead it to its snack. There are lots of ways to forage and lots of circumstances that change the way an organism will forage. For instance, a snake hunting in a stream will experience a dramatic temperature change when entering the water. This is important because it has been found that garter snakes are more likely to be active when their body temperature is at a preferred rate of greater than 30°C (Peterson, 1987). Also they will have to alter their buoyancy and change their breathing rate in order to hunt successfully (Lind, 1994).

            Enrichment studies are common in mammals. The effects of enrichment are easy to see in mammals because they react in a way that humans understand and are easier to test. A mouse that is housed in a cage with no stimulation is found to have impaired brain development, stereotypies which are or repetitive behaviors such as pacing or chewing, and anxiousness. These are all reduced when the mouse is given some sort of stimulus (Wolfer, 2004). Yet despite how important enrichment has been found to be for mammals, reptiles have not been tested for the effects of enrichment.

            There have been a few studies on the effect of enrichment on reptiles. One focused on performance in rat snakes. Almli and Burghardt (2006) tested performance of ratsnakes housed in enriched and unenriched environments. They used 16 ratsnakes and looked at how they performed during a problem solving task, a feeding task, and a field task. They found that the enriched snakes performed better than the unenriched snakes in the problem solving task and field task but in the feeding task there was no difference. A study on captive sea turtles found that enrichment is effective with marine reptiles (Therrien, 2007).

            In this experiment, I examined the role of environmental enrichment in the ability of the common garter snake (Thamnophis sirtalis) to find and capture a prey item. I hypothesized that the snakes reared in an enriched environment will find and capture their prey item faster than the snakes reared in an unenriched environment. If the enriched snakes are faster at discovery and capture of prey it can be concluded that enrichment does play a role in foraging behavior. If the unenriched snakes are faster at finding and capturing their prey, it can be concluded that enrichment inhibits the snake’s ability to find and capture their prey item.


Study Species

Eight young garter snakes, Thamnophis sirtalis, were obtained from Scott Felzer’s Garter Snakes, a private breeder in Raleigh, North Carolina. They were all roughly the same age and size and were raised in the same type of environment. We obtained three subspecies of garter snakes (T. s. sirtalis,    T. s. similis and T. s.radix). The garter snakes were all bred and raised in captivity but their native habitats were not identical. Four Florida Blue Striped Garter Snakes, T. s. similis, were received. They are native to Florida. One Eastern Garter Snake, T. s. sirtalis, was received. It is native to most of the eastern United States into the Midwest. Three Plains Garter Snakes, T. s. radix, were received. It is found over much of the Midwest. Four female and four male snakes were obtained.  The snakes were split up so that there were two males and two females in each group, and so that there was a similar mean body weight for each group. The groups were split up by subspecies as follows: enriched group had two T. s. similis, one T. s. sirtalis and one T. s. radix; control group had two T. s. similis and two T. s. radix.

Setting up the Housing

            Eight 10 gallon aquariums were used as everyday housing for the snakes and they were covered with metal mesh lids. Each snake had an overhead heat lamp that stayed on for 12 hours a day. They were lined with shredded coconut bark as the substrate and contained a large water dish.  The enriched habitats had additional items which included a broken ceramic pot, a small piece of driftwood and several medium sized rocks. The test environment consisted of a 25 gallon aquarium. They did not have a lid on them, because it would have obstructed viewing during the test. It was lined with a reptile mat. The reptile mat was a felt lining that is easily cleaned. In the test aquarium the exact same items were not used. Only the water bowl was the same across test and housing aquariums. In the test aquarium PVC pipe, flat large rocks, and a small hollowed out half of a log were used as items in the aquarium.  The test environment was set up with items in a different place than in the housing environment. The conformation of the test aquarium was changed each week. 


            The snakes were fed one large or two small minnows only once a week on test days. The minnows were obtained from local pet stores and maintained in a 10 gallon aquarium. The minnows were placed in the water bowl and the water bowl was placed at the end of an obstacle course. This not only discouraged habituation to a specific spot in the cage but it also forced the snake to use its instinctual foraging behavior to find its food.


            Each snake was tested once a week. Monday, Tuesday, Wednesday and Friday one control and one experimental snake was chosen for testing (Fig. 1 and 2).  The test aquarium was set up with its unique configuration, the minnow was put in the water dish in the aquarium, and then the snake was released into the aquarium. In order to minimize disturbance caused by the observation, a video camera was set up before the snake was released to capture all movement and behavior. The minnow was placed in a different place each time the snake was fed in the test aquarium. After each trial everything was taken out of the cage and wiped it down with a weak acetic acid mixture to remove all odors from the snakes. The length of time to food capture was recorded. The snakes were allowed two and half hours to find and capture the minnow. If they did not find and capture it within that time limit, they were left in the aquarium longer so they would eat the minnow. 15000 seconds was entered for the time to capture for the snakes that failed to find and capture the prey within two and a half hours.


            A two tailed T-test was performed to find significance of the data. SPSS 14.0 for windows was used to run the analysis.


Snakes Tested


1, 5


2, 6


3, 7


Text Box: Figure 2: This figure shows which snakes are control and which are experimental.


4, 8




Enriched snakes









Text Box: Figure 1: This was the rotation for the snakes. As you can see each snake was tested on the same day each week.







Environmental enrichment did not influence the foraging behavior of garter snakes. The average time to food capture for the enriched snakes (n=4) was 4659.11 seconds (SD=3058.72sec). The average time to food capture for the unenriched snakes (n=4) was 2164.65 seconds (SD=960.05sec). This difference in average did not translate into a significant difference (t=1.56, df=6, p=.171)(Chart 1).  In order to see if adding the value of 15000 seconds made a difference to the significance of the test we ran the data again with zero seconds in for the times that the snakes went longer than 2.5 hours. For this set of data the average time to food capture for the enriched snakes (n=4) was 2620.03 seconds (SD=1687.20sec). The average for unenriched snakes (n=4) was 1298.41 seconds (SD=541.72sec). This difference in average was not significant either (t=1.49, df=6, p=.186)(Chart 2).


 Text Box: Chart 1: This graph shows the average times to food capture for each group over the four weeks. Pulled snakes times=15000s.







Text Box: Chart 2: This graph shows the average times to food capture for each group over the four weeks. Pulled snakes times=0s.




The results of this experiment did not support the hypothesis that garter snakes (Thamnophis sirtalis) housed in an enriched environment would find and capture prey in a shorter period of time than a garter snake housed in an unenriched environment. The study suggests that enrichment does not have an effect upon the foraging behavior of garter snakes. This study only focused on a single behavior and a single way of performing enrichment. There are many other factors that need to be tested and controlled.

The hypothesis was not supported but this experiment still taught us a lot about the foraging behavior of garter snakes as well as the effects of enrichment. First off we see that human disturbance impacts these snakes in a variety of ways. Some snakes were more habituated to human presence and would calmly proceed to search their aquarium, but still keep a lookout for the human. Other snakes that were not as habituated to human disturbance would either quickly flee or posture aggressively and would seldom strike. This is a normal defensive strategy for garter snakes (Keogh, 258). A large predator would most likely just look over the snake and not see it if it stayed still and was in a place with lots of debris. A smaller predator would think twice about attacking a feisty little snake if it displayed aggressively and struck multiple times. That is an instinctual behavior.

These observations show that when testing a behavior in snakes, especially snakes that are not habituated to human presence, a video camera or other observation system should be used. Human presence should be limited to as little time as possible. Moving the snake and placing prey items should be the only disturbance. 

Temperature is a second variable that could have affected the foraging behavior. Temperature is a very important factor for a snake’s life. A garter snake is a poikilotherm, which means it is cold blooded. It cannot control its body temperature. A snake relies on its environment to raise its temperature. This rise in temperature allows the snake to digest its meal, move around freely, and could even affect its health and resistance to disease (Lueth, 125). The temperature in Voigt Science Hall during the fall semester of 2008 was not very constant. The system of temperature control is very inconsistent when the temperature fluctuates dramatically outside. The temperature of the building affected the temperature of the room that the snakes were kept in. Each snake was provided a heat lamp but that did not completely prevent temperature fluctuations. The temperature of their home habitats fluctuated between 72 and 87 degrees Celsius. This would affect the foraging behavior because the room might have been really warm some days and colder on other days. On the warm days the snakes would have been more willing to eat a large prey item while on cooler days they might not have desired to eat.

This variable can be controlled easily. The biggest hurdle to controlling it for this experiment was the inability to directly control the temperature that the snakes experienced. If this test was done in a lab setting with equipment such as incubators, or even a central system for a smaller section of the building, this variable could have been controlled.

Some other factors of snake behavior could have added variation to the results of this experiment. Shedding cycles, individual temperament, and general health could have all affected the snake’s food finding ability. A garter snake, as well as all other species of snake, shed their skin. The number of times that a snake sheds a year depends on its age. A young snake sheds more than a snake that is a few years old because the young snake is growing at a more rapid rate. Shedding is controlled by hormones and is associated with growth. The whole process of shedding can last up to three weeks. Shedding is preceded by a time of relative inactivity. This is because before a shed the scales covering the eye becomes opaque and hard to see through. Each of the tested snakes shed twice during the experiment. This means there was a period of relative inactivity for each snake during the experiment. The temperament of each snake could have caused slight variability for the results. The main expression of this temperament was observed during handling, and placing the snakes in the test aquarium. The temperamental snakes were the ones that would strike and get into a defensive posture. Young snakes tend to me more temperamental and not as habituated to human presence and handling.

The last factor that could have given variability to the results was general health. Two of the enriched snakes died during the eight weeks of the experiment.  Their deaths could have been caused by several things. They could have had a vitamin deficiency from such a narrow diet. A varied diet is best for snakes because it allows them to get all of the nutrients and vitamins that they need for survival. Another cause of their death could have been a parasitic or bacterial infection. They could have come in with a parasite or bacterial infection that didn’t was not noticed. All of these things could have caused health problems for the snakes that did not get noticed until they died or recovered. A slow snake or aggressive snake could have just been a sick snake which would have altered their behavior.

All of the variables above have shown that foraging behavior can be altered in many ways. Foraging is a behavior that is very instinctual and can be affected by many things. It is not a behavior that needs to be learned, especially in reptiles. Therefore, time to food capture might not have been the best behavior to test in this experiment. Some other behavior that might have worked better is time spent searching or exploring. This could have been done by a similar testing method but instead of focusing on the water bowl with the fish, the whole aquarium would have been in view of the camera. Then a grid put over top the picture to see how many squares the snake traveled through during a given period of time. Alternatively, one might test if the snakes had an aversion or affinity to new objects or substances. This could have been done in a similar way to the previous searching test. In addition to looking to see how many squares the snake traveled through, the amount of each substrate that the snake tested out and which substrate the snake spent most of its time on could be determined.

            This experiment did not give an answer to the question that we set out to investigate but it did give insight into the foraging behavior of garter snakes. It gave new ideas for future studies and gives new ways to go about it more productively. It did not answer the bigger question of whether reptiles need enrichment or not because it was very specific in the behavior tested whereas enrichment will affect many behaviors, but it did give an idea on what type of behaviors need to be tested.


Works Cited

Almli, Lynn M., and Gordon M. Burghardt. "Environmental Enrichment Alters the Behavioral Profile of Ratsnakes (Elaphe)." Journal of Applied Animal Welfare Science 9 (2006): 85-109.

Ford, Neil B. "Experimental Design in Studies of Snake Behavior." Herpetological Monographs 9 (1995): 130-39.

Keogh, J. Scott, and Frank P. Deserto. "? Temperature Dependent Defensive Behavior in Three Species of North American Colubrid Snakes." Journal of Herpetology 28 (1994): 258-61.

Lind, Amy J., and Hartwell H. Welsh, Jr. "Ontogenetic changes in foraging behavior and habitat use by the Oregon garter snake, Thamnophis atratus hydrophilus." Animal Behavior 48 (1994): 1261-273.

Lueth, Francis X. "Effects of Temperature on Snakes." Copeia 1941 (1941): 125-32.

Peterson, Charles R. "Daily Variation in the Body Temperatures of Free-Ranging Garter Snakes." Ecology 68 (1987): 160-69.

Wolfer, David P., Oxana Litvin, Morf, Nitsch, Lipp, and Wurbel. "Laboratory Animal Welfare: Cage Enrichment and Mouse Behaviour." Nature 432 (2004): 821-22.