Written by: Marissa Leonard

Revised by: Cassie Dutcher

Finalized by: Phong Los

       The fight or flight response is a unique and fascinating complex that is integral to the survival and continuation of many animal species, with the ultimate purpose of maximizing individual fitness (Engelhardt, et al, 2011). The organism must carefully weigh the costs and benefits of remaining when approached by a threat: retreating too soon may result in a waste of time and energy, while retreating too late could be potentially fatal (Kramer, et al, 1997). If the threat is ultimately deemed to be too close, and the potential risk of remaining is greater than the gain of staying in any particular area, the animal will take flight. This distance at which an individual flees from a potential threat (Engelhardt, et al, 2011) is called the Flight Initiation Distance (FID). In terms of the two respective experiments, the squirrel FID will be the distance a human can approach before the squirrel takes flight, flight being defined as running at least one meter away from the threat. The human FID will be the distance at which an unfamiliar person can approach the subject before the subject visibly moves away from the perceived threat. Escape behavior in humans shall be defined as moving/turning away or fleeing from the threat/perceived threat. Since the amygdala, a part of the human brain, cannot distinguish between a real threat or a perceived threat (VanItallie, 2002), the fight or flight response will be initiated automatically whenever the subject feels stressed. This response is caused by the autonomic nervous system’s responsibility of involuntary regulation of bodily functions (Gabella, 2001). In this scenario, the approaching stranger is the perceived threat.

       The initiation of the fight or flight response is a direct result of the stress hormone epinephrine (or adrenaline) being released in the body, thus resulting in physiological changes. These physiological changes are ultimately the result of the gene Phenylethanolamine N-Methyltransferase, or PNMT (Hoehe et al, 1992), which has been found in both humans and some species of squirrels. This gene, found on chromosome 17 in humans (Martin et al, 2001), codes for the protein, similarly named PNMT, that converts norepinephrine into epinephrine. The protein PNMT methylates the amine in norepinephrine by catalyzing the transfer of the methyl group from the cofactor S-adenosyl-L-methionine (SAMe) to norepinephrine (Wong, 1992), thus creating epinephrine. This reaction occurs as a response to stress hormone levels or the amount of nerve impulses due to stress. Therefore, if an organism is experiencing higher levels of stress, more epinephrine is produced (Wurtman, 2002). The action of epinephrine does not exert negative feedback to down-regulate synthesis, and instead is terminated by the reuptake into terminal nerve endings (Martin et al, 2001).

       In Homo sapiens, the amygdala first sends a signal to the hypothalamus, which then activates the pituitary gland into releasing adrenocorticotropic hormone, or ACTH (Margioris et al, 2011). The release of ACTH then activates the adrenal glands,- located above the kidneys- which then begins to produce epinephrine. Epinephrine then binds to receptors on the liver and muscle cells, thus resulting in a release of glucose from stores in the liver and muscles, (Flatt et al, 1964) thus energizing the person. Bodily physiological changes include increased blood sugar and blood pressure, accelerated heart and lung rate, dilation of blood vessels in muscles. (Hoehe et al, 1992) It has been proven that Citellus citellus, European ground squirrels, contain the PNMT gene (Petrovic et al, 1977) which codes for the PNMT protein, thus playing a key role in the regulation of epinephrine production. Therefore, it is reasonable to assume that the squirrel specimen studied in this experiment, Sciurus niger, the fox squirrel, may also contain the PNMT gene. Since PNMT is considered to be a catecholamine, it plays a significant role in responding to a perceived threat and in augmenting cardiac function in squirrels as well (Kvetnansky, 2004).

       To investigate the difference in FIDs between eating and non-eating organisms, we closely followed the methods of S.C. Engelhardt’s 2011 study, “Effects of levels of human exposure on flight initiation distance and distance to refuge in foraging eastern gray squirrels (Sciurus carolinensis)”. Though we studied the fox squirrel (Sciurus niger), instead of the eastern gray squirrel (Sciurus carolinensis), the two organisms are related enough in that the methods can be replicated. Since the predictions and hypotheses are the same for our human and squirrel experiment, to continue to ensure consistency, we utilized Engelhardt’s methodology for the human experiment as well. In Engelhardt’s experiment, the experimenters themselves approached the squirrels until they fled, and then measured the distance between the approximate location of the squirrel and used the final location of the experimenter. However, in attempt to minimize bias, we did not approach the test organisms ourselves, because this could have potentially influenced squirrel/human behavior in a way that would skew the results. By making observations instead of purposefully influencing behavior, we will be able to determine actual FIDs while minimizing potential bias.

       In this experiment, we are investigating if the fight or flight response significantly affects the distance at which a threat/perceived threat can approach the subject while they are actively eating, as compared to when they aren’t eating, before they exhibit escape behavior (FID). Using Michigan State University students and Michigan State University campus squirrels as our model organisms, we plan to investigate the effects of actively eating on FID. For this experiment, “actively eating” shall be defined as 1. (for squirrels) Containing food in its mouth or engaging with foodstuffs on the ground; 2. (for people) Interacting with the food on the plate or containing food within their mouth. We hypothesize that squirrels and humans will exhibit escape behavior when the cost of remaining is higher than the cost of fleeing, and that eating will increase the cost of fleeing due to its relation to survival (Cooper et al, 2007). We plan to test the following predictions: 1. Squirrels who are actively eating will have a lower FID than squirrels that are not eating; 2. MSU students who are actively eating will have a lower FID and be more reluctant to exhibit escape behavior than students who are not eating (Lovallo, et al, 2000). We predict a lower relative FID in both organisms because eating activates the parasympathetic nervous system, which is the biological opposite of the fight or flight response, which activates the sympathetic nervous system (Lovallo, et al, 2000). However, since our study focused specifically on the effects of feeding in FID, we anticipated that the parasympathetic nervous system would cause a delayed fight or flight response attributed to the regulation of energy intake through feeding and digestion (Cooper et al, 2007).

Written by: Phong Los

Revised by: Marissa Leonard

Finalized by: Cassie Dutcher

       To begin collecting observational data for the Sciurus niger, or fox squirrel, test sites were first chosen. Michigan State University’s intercampus population density remains relatively constant in the number of people throughout its developed acreage. Due to this uniformity, any site located within the geographical radius (2,000 developed-acres) of MSU’s East Lansing campus was able to be utilized for experimentation. The most common observation sites were near Holmes Hall, Wells Hall, and the main library, though data was collected all over MSU’s campus. Data was collected from both male and female fox squirrels. As stated above, eating is defined as the squirrel clearly containing food in its mouth (acorns, nuts, etc.) Non-eating is defined as the opposite- clearly not consuming any food or engaging with food in any way. Throughout the experiment, data was only recorded if humans acted as the perceived threat to squirrels, meaning that they were initiating the squirrels’ flight response. A perceived threat is defined in this study as a human approaching close enough to the squirrel so that the squirrel moves away from said human. The two treatments in this study are the squirrels eating or not eating. Since this is an observational study, there are no controls in this experiment.

       We observed squirrel-human interactions from at least 3 meters away as not to interfere with the behavior of the squirrels or influence the resulting FID, which was done to reduce bias (Engelhardt, et al, 2011). The squirrel in question was categorized into one of the two previously stated categories of either feeding or nonfeeding, and the interaction was observed. This interaction between squirrel and human was recorded digitally with an iPhone 6 camera, and observations were written in a lab notebook. Once the interaction had taken place and the squirrel had fled, the FID was determined and then recorded. If a squirrel fled because of a human threat, resulting in a quantifiable FID, the approximate distances between human threat and squirrel were immediately determined. The FID was obtained by placing a flag marker at the initial position of the squirrel, and another flag marker at the closest distance that the threat had approached the squirrel before it ran off. The distance between the two flags was then measured with an expandable tape measure and then recorded in a notebook (Engelhardt, et al, 2011). This was done by approximation of their locations as seen by the observer. The sample size (n) for total observations completed was at least 30 (Dill et al 1989) so that there was a large enough sample size from which statistical significance could be determined.

       SA statistical analysis was then performed on the data using a paired t-test for the difference in sample means. This particular test was used to determine if the difference in the two mean FIDs were statistically significant. We also had to meet a series of conditions in order to use this test to analyze our data: 1. Simple random sampling must be used; 2. The samples must be independent from one another; 3. The population must be 20 times larger than our sample size; 4. The sampling population is approximately normal. For our experiment, we assumed that we met all of the aforementioned criteria, and therefore proceeded with the statistical analysis. Two separate t-tests were conducted, so there was one statistical test performed for each organism. The test statistic was obtained using the following equation, where x̄=mean FID, N=sample size, and S=sample standard deviation:

       For the statistical test performed in this experiment, null and alternate hypotheses were formed. The null hypothesis for each test is H₀ : x̄_eating= x̄_{non eating}, which means that the sample mean FIDs for eating and non eating squirrels will be roughly the same. The alternative hypothesis for both tests is H_a : x̄_eating < x̄_{non eating}, meaning that the mean FID for eating squirrels will be less than the mean FID for non eating squirrels. The alternative hypothesis purposely reflects our prediction that the mean FID for eating squirrels will be lower than the mean FID for non eating squirrels. By finding the t-value of the test statistic, we can then determine the p-value, and the resulting p-value is then compared to a standard alpha value of α=0.05. If the p-value is greater than 0.05 we fail to reject the null hypothesis and therefore, cannot assume that the mean FID is different between eating and non-eating squirrels. If the p-value is less than 0.05, we reject the null hypothesis and can assume statistical significance in that the mean FID for eating squirrels is significantly lower than the mean FID for non-eating squirrels.

Written by: Marissa Leonard

Revised by: Phong Los

Finalized by: Cassie Dutcher

       The test site utilized in this experiment was the 2,000 developed acres of MSU’s East Lansing campus. Since the population density on campus is relatively uniform, so we were able to utilize all areas for observation in the human experiment as well. Specific data points were collected primarily at the cafeterias, where food was present, although additional data collection locations included the main library, and at the multiple “Sparty’s” convenience stores located throughout campus. Data was collected from both male and female students of all ages. The two treatments in this study are the squirrels eating or not eating. Since this is an observational study, there are no controls in this experiment.

       Eating was defined as a person interacting with food on a plate or containing food in either their mouth or hand. Not eating was classified as exhibiting none of the “eating” behaviors. Since humans do not always flee from a stressful situation, we would record an FID if subjects exhibited escape behavior in response to the approach of a clearly unfamiliar person, considered a “perceived threat”. Escape behavior is defined as the either exhibiting nonverbal signals expressing introversion, creating space between them and the perceived threat, leaving the situation all together (fleeing), or a combination of all three. Introverted nonverbal signals and gestures in humans include backward leaning, turning away, an inward direction, and a slow response time (Neff, et al, 2010). A perceived threat is classified as a person who, though not necessarily posing an actual threat to the subject, is clearly unfamiliar to the subject in question and may cause the subject to undergo stress.

       The same statistical test (paired t-test for the difference in sample means) was utilized for the analysis of our Homo sapien data. The null and alternative hypotheses remain the same from the squirrel experiment to test the mean FIDs of eating and non-eating humans: H₀ : x̄_eating = x̄_{non eating} and H_a : x̄_eating < x̄_{non eating}. The same formula to calculate the test statistic t was utilized (Homo sapien Methods), from which the p-value was then determined. The calculated p-value was then compared to a standard alpha value of α=0.05. If the p-value is greater than 0.05 we fail to reject the null hypothesis and therefore, cannot assume that the mean FID is different between eating and non-eating humans. If the p-value is less than 0.05, we reject the null hypothesis and can assume statistical significance and infer that the mean FID for eating humans is significantly lower than the mean FID for non-eating humans.

Written by: Phong Los

Revised by: Marissa Leonard

Finalized by: Marissa Leonard

       We predicted that Michigan State University fox squirrels would have an observed lower FID when eating, compared to when they are not eating (Figure 1). Essentially, we predicted that when a squirrel is eating, the threat should be able to get closer to the squirrel before it takes flight than they would if the squirrel wasn’t eating, resulting in a lower FID. Contrastingly, when an observed squirrel is not eating and is approached by a human threat, they would have a larger FID. This is because eating utilizes the parasympathetic nervous system, which is the biological opposite of the fight or flight response, which invokes the sympathetic nervous system (Lovallo, et al, 2000). Therefore, we predict that it will take longer to transition from opposing biological states when squirrels are more relaxed, and therefore the FID will be lower than when not eating.

Written by: Cassie Dutcher

Revised by: Phong Los

Finalized by: Marissa Leonard

       We predicted that students on Michigan State University’s campus who are eating would display a shorter FID in comparison to students who are not eating. This is because people who are multi-tasking are more distracted, and are therefore less aware of their surroundings and environmental influences (Ophir, et al, 2009). The multi-tasking factor in our experiment is the treatment of eating versus not eating. Therefore, we predicted that this distraction would allow a perceived threat to get closer to a distracted subject without the subject displaying escape behavior. Additionally, since humans also have a sympathetic and parasympathetic nervous system much like their squirrel counterparts, we hypothesize that it will take longer to transition into a state of stress when humans are initially relaxed (Lovallo, 2000). Because of this evidence, we predict the FID due to a human threat will descend in increasing order of the mean FID (Figure 2).

Written by: Phong Los

Revised by: Cassie Dutcher

Finalized by: Marissa Leonard

       Once the Sciurus niger trials were completed, the means for both eating and non-eating squirrels were calculated from the number of observations completed in the total sample of thirty. For eating squirrels, the following numerical values were calculated: x̄=1.51 meters ; N = 17 ; S = 1.65 meters. For non-eating squirrels, the following numerical values were calculated: x̄=3.87 meters ; N = 14 ; S = 1.37 meters. A paired t-test for the difference in sample means was then ran and we obtained a t-value of -4.35 and a p-value of 7.80x10^-4. Since the calculated p-value is less than the standard, α=0.05, we can reject the null hypothesis (H₀ : x̄_eating= x̄_{non eating} ) and conclude that x̄_eating < x̄_{non eating}. This simply means that there is statistical significance between the two mean FIDs, and that the mean FID for eating squirrels was lower than the mean FID for non-eating squirrels, thus supporting our initial prediction (Figure 3A).

       Similar to the squirrel results, the means for both eating and non-eating homo sapiens were calculated from the total number of observations in the total sample size of 31. For eating humans, the following numerical values were calculated: x̄=0.73 meters ; N = 16 ; S = 0.423 meters. For non-eating humans, the following numerical values were calculated: x̄=1.15 meters ; N = 15 ; S = 1.27 meters. Another paired t-test for the difference in sample means was used to analyze the data for human FID. The test statistic, or t-value, was calculated to be -1.21 and the resulting p-value was calculated to be 0.12. Since the calculated p-value is greater than the standard, α=0.05, we fail reject the null hypothesis (H₀ : x̄_eating= x̄_{non eating}) and can conclude that: x̄_eating < x̄_{non eating}. Since we have failed to reject the null hypothesis, we assume that there is no statistical difference between the the mean FIDs of eating and non-eating humans (Figure 3B).

       Boonstra, R., C. J. McColl, and T. J. Karels.2001. Reproduction at all costs: The adaptive stress response of male arctic ground squirrels. Ecology. 82: 1930-1946.

       Cooper, W. E., W. G. Frederick. 2007. Optimal Flight Initiation Distance. Journal of Theoretical Biology. 244: 59-67.

       Dill, L., and R. Houtman. 1989. The influence of distance to refuge on flight initiation distance in the gray squirrel (Sciurus carolinensis). Canadian Journal of Zoology. 67(1): 233-235.

       Engelhardt, S.C. R.B. Weladji. 2011. Effects of levels of human exposure on flight initiation distance and distance and distance to refuge in foraging eastern gray squirrels (Sciurus carolinensis). NRC Research Press. 89: 823-830.

       Flatt, J. P., E. G. Ball. 1964. Studies on the metabolism of adipose tissue: An evaluation of the major pathways of glucose catabolism as influenced by insulin and epinephrine. The Journal of Biological Chemistry. 239: 675-685.

       Gabella, G. 2001. Autonomic Nervous System. Encyclopedia of Life Science. 1: 1-7.

       Hoehe, M.R, R. Plaetke, B. Otterud, D. Stauffer, J. Holik, W. F. Byerley, E. E. Baetge, E. S. Gershon, JM. Lalouel, and M. Leppert.1992. Genetic linkage of the human gene for phenylethanol-amine N-methyltransferase (PNMT), the adrenaline-synthesizing enzyme, to DNA markers on chromosome 17q21. Human Molecular Genetics. 1 (3): 175-178.

       Kramer, D.L. M. Bonenfant. 1997. Direction of predator approach and the decision to flee to refuge. US National Library of Medicine, National Institutes of Health. 54(2): 289-95.

       Kvetnansky, R., L. Micutkova, L. Kubovcakova, E. L. Sabban, M. Palkovits, and O. Krizanova. 2004. Localization and Regulation of Phenylethanolamine N-Methyltransferase Gene Expression in the Heart of Rats and Mice during Stress. Annals of the New York Academy of Sciences. 1018: 405-417.

       Lovallo, W.R., J. J. Sollers III. 2000. Autonomic Nervous System. Encyclopedia of Stress. 3: 275-277.

       Margioris, A. N., C. Tsatsanis. 2011. ACTH Action on the Adrenal. Endotext [Internet]. South Dartmouth, MA.

       Martin, J. L., J. Begun, M. J. McLeish, J. M. Caine, G. L. Grunewald. 2001. Getting the Adrenaline Going: Crystal Structure of the Adrenaline-Synthesizing Enzyme PNMT. Structure. 9: 977-985.

       Neff, M., Y. Wang, R. Abbott, M. Walker. 2010. Evaluating the Effect of Gesture and Language on Personality Perception in Controversial Agents. Intelligent Virtual Agents; Lecture Notes in Computer Science. 6356: 222-235.

       Ophir, E., C. Nass, A.D. Wagner, M. I. Posner. 2009. Cognitive Control of Media Multitaskers. PNAS. 106(37): 15583-15587.

       Petrović, V. M., and O. Rajčić. "Accelerated gluconeogenic processes in the ground squirrel (Citellus citellus) during the arousal from hibernation."Comparative Biochemistry and Physiology Part A: Physiology 80.4 (1985): 477-480.

       VanItallie, T.B. 2002. Stress: A Risk Factor For Serious Illness. Metabolism. 51: 40-45.

       Wurtman, R. 2002. Stress and the Adrenocortical Control of Epinephrine Synthesis. Metabolism: 51: 11–4.

       Wong, D.L., A. Lesage, B. Siddal, J. W. Funder. 1992. Glucocorticoid Regulation of Phenylethanolamine N-Methyltransferase in Vivo. The FASEB Journal. 6: 3310-3315.

Written by: Phong Los

Revised by: Cassie Dutcher

Finalized by: Marissa Leonard

Written by: Phong Los

Revised by: Marissa Leonard

Finalized by: Marissa Leonard

       Figure 1: Predicted relative mean Flight Initiation Distance (FID) in feeding and non-feeding fox squirrels (Sciurus niger). The figure depicts our predicted data for the squirrel observational study. The blue bar represents actively feeding squirrels and the red bar represents the non-feeding squirrels. We predicted that it will take longer for the Sciurus niger to transition from opposing biological states, -from a state of relaxation into a state of stress- and therefore the FID for eating squirrels will be lower than non-feeding squirrels (Lovallo, 2000). No true data was used in the figure, only a predictive trend is shown.

Written by: Cassie Dutcher

Revised by: Phong Los

Finalized by: Marissa Leonard

       Figure 2: Predicted relative mean Flight Initiation Distance (FID) of eating and not eating humans (Homo sapien). The figure depicts our predicted data for the human observational study. The blue bar represents actively eating humans and the red bar represents the not eating humans. We predicted that the mean FID will be lower for eating humans than it would be for non-eating humans, since people who are multi-tasking ,which can include eating, are more distracted, and are therefore less aware of their surroundings and environmental influences (Ophir, et al, 2009). No true data was used in the figure, only a predictive trend is shown.

       Figure 3: Mean Flight Initiation Distance (FID) and standard errors (SE) of fox squirrels and humans in response to a threat while feeding or non-feeding. Part A depicts the data obtained from the squirrel (Sciurus niger) experiment. For feeding squirrels: x̄=1.51 meters ; SE = ±0.40. For non-feeding squirrels: x̄=3.87 meters ; SE = ±0.37. The measurements of FID were obtained by placing flag markers at the location of the threat and the location where the squirrel exhibited FID, and then measuring the distance between the two flags in meters. The y-axis depicts the mean FID in meters in relation to the x-axis portraying whether the organism was feeding or not. Using a paired t-test for the difference in sample means it was determined that the means for feeding and non-feeding squirrels were significantly different (P < 0.05). Part B depicts the data obtained from the human (Homo sapien) experiment. For eating humans: x̄=0.73 meters ; SE = ±0.11. For non-eating humans: x̄=1.15 meters ; SE = ±0.32. The FIDs of actively eating or not eating humans were measured and recorded above using the same methods to obtain the squirrel data. The y-axis shows mean FID of humans in meters in relation to the x-axis illustrating whether the human was eating or not eating. By completing a paired t-test, the difference in mean FIDs for eating vs. not eating humans was not found to be significant (P > 0.05).