PRESENTATION:

BACKGROUND:

Hermit Crab Stimulus Concordance and Discordance

In order to survive and reproduce, animals will use their cognition, defined as their retention, acquisition, and use of information to avoid predators and find potential mates (Stahlman et al. 2010). Specifically, Stalhman et al. states that animals rely on producing or receiving biologically important sounds such as calling for and recognizing mates, spatial navigation such as echolocation in bats, and detecting noises unique for hunting prey and avoiding predators (2010). To avoid being killed by a predator, prey will determine how dangerous a predator is by how far away they are (Hingham et al., 2015). This determination is a continuum between dying or, if the prey feels safe, choosing a nesting site (Yim-Hol Chan et al., 2010 and Putman et al., 2015). An animal's ultimate goal, when faced with a possible predator is to avoid any damage to their tissues caused by possible noxious stimuli (Magee et. al, 2016). The most common physiological responses reported in crustaceans is hiding, running away, or a combination of the two (Tomsic et. al, 2017). For example, Neohelice crabs protect themselves from predators by running to an individual burrow where they hide if they assess the risk is great enough (Fathala and Maldonado, 2011). A crab’s response time is constantly adjusted based on the visual information obtained by the animal (Tomsic 2017). For example, crabs will accelerate when they’ve determined an area is dangerous and decelerate after they’ve received sufficient cues to determine the new location is safe (Tomsic 2017). Chemosensory cues or semiochemicals are also an important form of predator detection due to its increased spatial range, persistence, and reliability (Wiseden and Chivers, 2006). These cues have the potential to contain a wider array of information compared to auditory cues such as current and recent physical closeness of predators, predator diet, and activity level. The strength of a chemical cue reflects the immediate threat level and are left by predators through the involuntary release of odors and metabolites known as kairomones (Katz and Dill, 1998).

Many predator interactions will require prey to multitask and process cues from both their environment and predator. Hallmarks of intelligent behavior include being able to multitask while focusing on one spatial location to achieve one’s goals (Golob, 2020). Extraneous noises can affect organism communication, change settlement decisions, or cause distractions that can increase vulnerability to predators. Humans and monkeys often focus their attention on what is most important to them, such as social interactions or looking for food, with their brain filtering out less important information (Dukas and Kamil, 2000). Distraction has been well documented in primates and rats, which can affect the response to visual threat (Chan et. al, 2010). Being distracted is often caused by limited attention spans and can influence an animal’s ability to respond properly to a threat, making it easier for predators to do lasting damage. (Chan et. al, 2010; Smith, 2010). This relationship between distractors and predator cues is often referred to as “sensory pollution” or “info-disruption” because of its negative effect on an animals’ ability to appropriately respond to stimuli (Hafwerk and Stabbekorn, 2015). In addition, these sensory effects can cause physiological stress (eg. avoiding certain areas), possibly altering behavior further (Kight and Swaddle, 2011).

Crustaceans, such as the hermit crab, have eyes that are mounted on two movable eye stalks which contain only one or two photoreceptor types, that make them sensitive to blue and green colors (Ping et. al, 2015). A number of studies suggest that the hermit crab (Paguroidea) may also have sophisticated visual abilities. One hermit crab species, Pagurus bernhardus, has been found to avoid shells that have the greatest contrast with the background, and choose shells that best match the color of their surroundings (Briffa et al., 2008). Another hermit crab, Clibanarius vittatus, can discriminate between different geometric shapes with equal surface area, and are also found to be more attracted to horizontal rectangles and less attracted to vertical diamonds (Diaz et al., 1994). Ping et. al suggests that the eyes of hermit crabs are most comparable to insect eyes due to their compound nature, meaning they have thousands of tiny units of cornea, lens, and photoreceptor cells to distinguish brightness and color.

Previous studies have shown that distractions can affect movement by slowing it down resulting in longer response times (Luo et. al, 2015). One example would be Chan el. al’s study in which white noise was presented alone for 10 seconds, followed immediately by the presentation of the same noise combined with an image of a predator. In another trial, the white noise was presented alone for 90 s prior to presenting the compound stimulus. The 90 s white noise trial produced a longer latency to hide (10 s) than the trial presenting the white noise for a shorter 10 s period. This difference is likely explained by the long-duration stimulus being more distracting than the short-duration stimulus, thereby attenuating the withdrawal response (Chan et al., 2010). However, Briffa et al. demonstrated that hermit crabs can exhibit individual variation in their behavior where one hermit crab may respond differently to a predator than another. This was investigated by testing a new assay for making such comparisons and investigating adaptability to different situations and consistency in the duration of predatory responses in the European hermit crab Pagurus bernhardus’ startle response in relation to distance. In the Briffa et al. study, they concluded that hermit crabs of different shell sizes had different levels of shell occupation when exposed to anthropogenic noise, indicating for the first time the presence of animal personalities in a crustacean that could be biological (Briffa et al., 2008). Therefore they concluded that variation in responses among individuals exposed to the same stimulus can result from differences in size, condition, and energy reserves (Houston and Macnamara, 1999). Because of this difference in individual behavior, one hermit crab may go about its life more cautious than another and may exhibit more of what we would categorize as fear, which is a reaction to a specific threatening stimulus. This behavior can result in increased avoidance to fear related cues that are often unreasonable compared to the average organism (Gorman et al., 2018 and Broeren et al., 2011).

Hermit Crabs often rely on their sound and sight to determine if they are at risk of harm from a predator through fear related cues. They are a good behavioral model organism because they can be easily monitored and have a measurable antipredator response, which is hiding in their shells (Hol-Chan et. al, 2010). This response has been heavily studied and is unique to animals with shells (Nanninga 2020). Hermit crabs may also exhibit freezing responses, which are also easy to detect and measure. The binary nature of a hermit crab’s predator response, makes them a useful model organism for measuring responses to predatory stimuli (Shragai et al., 2017). Hermit crab behavior has been heavily studied because their taxon is known to exhibit highly variable responses, even between individuals of the same species (Gherardi et al., 2012). According to Gherardi et. al, this implies that hermit crabs are able to exhibit a wider array of responses, which indicates higher intelligence and an ability to process multiple, complex environmental stimuli (2012). In general, crab species are especially sensitive and respond to a lower threshold of stimuli when compared to many other species, making it easier to replicate a predatory environment (Gruber et al., 2019). Although extremely sensitive to all stimuli, when exposed to a non-predictive noise, hermit crabs are slower to withdraw back into their shells by about 2 seconds. This finding supports that crabs process auditory stimuli and make informed behaviors rather than responding strongly to all stimuli (Ryan et. al, 2012). One knowledge gap is whether hermit crabs can hear and process sounds, but, due to a recent study by Tomsic et al., it’s clear that they are able to detect low to mid frequencies of about 5 nm to 100 nm of acoustic noise (Tomsic et al., 2017). While there is less information on hermit crab hearing, the visual system has been more heavily studied.

For the first iteration of my experiment, I replicated the experiment of Ryan et al. in which nine hermit crabs were exposed to three different trials of auditory and visual predatory stimuli (Ryan et al., 2011). In each trial, one hermit crab was clamped to a c-clamp in a quiet and dark testing room. Once they were clamped in, I waited for the crab to come out of its shell for thirty seconds straight, and then started playing a video on a monitor that was 25 cm away from the hermit crab. The video I used was of an image of a coconut crab that slowly got bigger for thirty seconds with white noise playing as well. There was a control trial with no sound, a concordant trial with the sound coming from the same direction as the video, and a discordant trial with the sound coming to the left of the crab. Then, after the video started playing, the time until the hermit crab retreated back into its shell was recorded. However there was only one successful trial where latency of hiding could be measured. Instead of hiding, most crabs became more tense and tried to escape from the c-clamp. This preliminary data indicates that little is known about what stimuli elicit hiding versus escaping behavior. From this work, it was clear that the dependent variable needed to be adjusted to include escaping behavior as well as freezing and hiding.

This current study aims to investigate how prey process predator cues to ensure their survival. Prior literature has demonstrated that larger spatial distances between stimuli cause prey to have a slower response to predatory cues. When the stimuli were projected from opposing directions, the authors noticed the slowest response indicating the prey were disoriented by the opposing cues. This is not perhaps surprising given that prey have to process the distance between themselves and a predator to successfully survive (Ryan et. al 2012). I hypothesize that, if we vary the direction between an auditory and visual stimulus, hermit crabs will exhibit different latency times before freezing in response to the stimulus, with the fastest response being the concordant trial. This behavior is due to anti-predator withdrawal, which slows their response time. To adapt my project to incorporate escape behavior, I will utilize Stahlman et. al’s experimental set up and, instead of using the c-clamp, the hermit crabs will be placed in an aquarium for testing. This will remove the challenge of securing crabs to the c-clamp and allow more freedom for the crabs to display a wider variety of response behaviors. The same concordant, discordant, and control trials will be used except that, within the testing tank, there will be a clear container in which the hermit crabs are forced to engage directly with the stimuli, and a touch stimulus will be adde to ensure that the crab hides in its shell. This experimental setup offers a unique opportunity to investigate all the facets of a predator/prey interaction in the controlled setting of a research lab.

One knowledge gap that may limit my research is whether the hermit crabs will hide when exposed to the stimuli. As mentioned earlier, when exposed to a predator hermit crabs will either attempt to run away or hide in their shells (Tomsic et al., 2017). While Chan et al.’s experiment successfully measured how long until the hermit crabs hid in their shells after being exposed to the stimuli, most of my experimental trials recorded data where the hermit crabs did the alternative, which was trying to escape and run away. As Yim-Hol Chan, 2010 determined, this may have to do with the relationship between the distance of the prey and predator, possibly signaling that the trials may have been more successful if the predatory stimulus wasn’t so close to the prey (Yim-Hol Chan, 2010).

This type of research has the potential to offer much needed insight into how prey process predator cues. This has become increasingly important as anthropogenic noise pollution has increased, which could cause unintentional distractions that limit prey to process and evade predator cues successfully (Gorman, 2018). Thus, the introduction of excessive anthropogenic noise can affect hermit crab survival, possibly leading to decreased fitness and extinction of the species (Ryan et al., 2012). The aforementioned relation between stimuli and distraction is applicable to all animals that rely on their attention abilities to determine their safety from predators (Chan et al., 2010). This information on the effects of distractions during sensing predator cues can also be applied to foraging behavior, another integral part of animal fitness. Foraging distractions have been tested in bird species, such as Blue Jays, who divide their attention between antipredator responses and hunting for food, which is essential for their survival (Ryan et al., 2012). If a future study was conducted, more directions between the audio and visual stimuli should be tested to determine whether specific directional differences, such as left vs right vs behind the crab, are more distracting than others.

METHODS & PROCEDURE:

Methods

Subjects and Care

9 Coenobita clypeatus, land hermit crabs, were purchased from Carolina Biological (Product 142396) along with a hermit crab care kit (Product 142396) which consisted of a 10 gallon aquarium, 250 g of dry hermit crab food, 10 lbs of white quartz crystal sand and heaters. The tank was set up so the sand covered the bottom of the tank, which prevented the crabs from slipping and allowed them to burrow. The crabs were given 2 petri dishes for water, which were replaced everyday, and another petri dish for food. Each crab was given a number 1-9, which was then written on their shell in Sharpie to make it easy to differentiate between them. Hands were washed with soap and water before and after handling the crabs. The crabs were also given new sand substrate every two months and fecal matter was scooped out weekly using a sieve. The crabs should be handled very delicately and with care.

Preliminary Study

In a preliminary study, 9 Coenobita clypeatus, (Land hermit crabs, Item #142369) were exposed to auditory and visual predatory stimuli. The visual stimuli consisted of a slowly increasing image of a coconut crab as used in Ryan et al.’s study (2012). The stimuli were recorded using the Video Leap video maker app for iPhone, where the image appeared at the top left corner of the screen and slowly increased for 30 seconds. Then, the auditory stimulus, which was white noise (Youtube link), was recorded over the video, which in total ran for 90 seconds (Maes & De Groot 2003). The visual stimulus was shown on a 17 inch LG monitor (model #‎24M47VQ) while the auditory stimulus was played on Logitech Z130 speakers. The hermit crabs were individually presented with three different predatory situations modeled after Ryan et al.’s study: a control trial where just the auditory stimulus was played, a concordant trial where the visual and auditory stimuli were played from the same direction, and a discordant trial where the speakers playing the auditory stimulus were placed 25 cm to the left side of the crab (2012). The preliminary study differed from Ryan et al.’s in that less hermit crabs were used and different pictures and audio were used to create the stimuli, which had no effect as the photo was of the same animal and there was still enough data points. For each trial, the audio was played at 89 db, measured using a sound level meter from Meterk (item #E2451), because Salmon et al’s study on crab auditory physiological systems found that this level of sound is most effectively heard by crabs (1971). In order to keep the hermit crabs in place, each crab was screwed into a c-clamp with their shell opening facing the monitor as seen in Ryan et al’s (2012) study and Nolan et. al (2004). The c-clamp proved problematic, as it was not an effective way to keep the hermit crabs in one place when exposed to the stimuli. As soon as the crab was screwed in place, a timer was started if they came out of their shells. After the crab was out of its shell for 30 seconds and allowed to acclimate, the video was started with the audio, if necessary for the given trial (O’Brien & Dunlap, 1975). Once the video was started, if the hermit crab went back in its shell, the timer was stopped. This time was recorded and measured as the hiding latency, which was used in Chan et al.’s study to demonstrate that extraneous noise negatively impacts anti-predator withdrawal behavior (2010). Although hiding was the expected response, many hermit crabs exhibited other typical fear responses cited by Tomsic et al., which included running away from the video because the c-clamp could not hold them adequately. This made it very difficult to record hiding latency data and many trials had to be repeated until the desired response was exhibited. This preliminary work showed that the hermit crabs could have a reaction to the stimuli presented but proved the holding method of the c-clamp needed to be changed as it didn't adequately hold the hermit crabs in place.

Final Study

For the final study, a testing arena was built based off of an experiment done by Wale et al., where a 10 gallon tank was used as the testing arena for the crabs. The glass at the bottom of the tank was covered with insulation pads (Item # 142396) to prevent extraneous vibrations against the glass of the tank. A 10 cm x 26 cm section of the tank was then marked off with a sharpie at the bottom right side of the tank. The 26 cm side was the short side of the bottom of the tank as shown below. A mesh screen (ordered from Amazon) was then cut, with scissors, inside the tank and lined up along the 10cm line and hot glued to the walls (Amazon). Fish tank gravel (from Amazon) was put in the larger portion of the tank and was used instead of the normal hermit crab sand so that the material used wouldn’t get in the holes of the speaker. To replace the c-clamp from the study, a see-through plastic 16 oz container was used to hold the hermit crabs in place similar to the one used in Purser & Radford’s study on attention shifts in three-spined stickleback fish (2011). Purser & Radford’s plastic container allowed the fish to be separated individually from each other in the tank, and was translated to my study so that each hermit crab could be well contained while also being able to see the visual stimulus, which was ineffective in the preliminary study. To allow the hermits to breathe and hear the noise, a hole 7 cm in diameter was cut at the bottom of the container. During the trials, the container was flipped upside down so that the hole was at the top. This enclosure ensured that the hermit crabs would be kept in one location without being able to escape for the duration of the trial.

The experimental setup was similar to the preliminary study, with the same trial conditions: concordant, discordant and control (Ryan et al, 2012). The same video was used, except it was shown on an 24cm x 17cm iPad (model #MR7G2LL/A) instead of a monitor to adapt to the small size of the tank. To make it easy to start and stop the video, a bluetooth Apple mouse (model #A1657) was used to control what was being shown on the iPad. A different smaller speaker, the Oontz angle (from Amazon), was used to play the white noise as it fit in the tank behind the mesh screen (as shown below) and had wireless bluetooth features. The hermit crab was then placed in the plastic cup with the visual and auditory stimuli ready to play. A timer was set and once the crab was out of its shell for 30 seconds and acclimated, the visual and auditory stimuli was started using the Apple mouse. After 20 seconds of the video playing, the hermit crab would be lightly tapped on its shell with a glass stir rod until it hid. Once the hermit crab came back out its shell after being tapped, the timer was stopped. The time was then recorded in the data table below under which concordant, discordant, or control trial it was. Each trial was done a total of nine times using each of the 9 hermit crabs (ie. 9 biological replicates tested once). The tapping ensured that all the crabs hid in their shells as a reaction to the stimuli. The idea is the video and touch stay the same but only the direction of the audio changes. This was done to address the varied types of fear responses seen in the preliminary study. Thus, the addition of the tap helped the dependent variable be a more consistent and measurable behavior as it eliminated other fear responses. (Elwood et al. 1998). Once each trial was finished, the average time for each trial type was measured and compared using a student T test assuming independent variables. At the end of the experiment, all data recorded was entered into various graphs and were determined to be outliers if they were about +- 30 seconds away from the average. Based on that calculation, trials were repeated if necessary. The p values were used to measure the statistical significance of the data . One limitation that occurred was that it was possible that the hermits learned and got used to the situations presented, causing them to not be affected by the stimuli presented.

RESULTS AND DISCUSSION:

In this study, nine land hermit crabs (Coenobita clypeatus) were presented with three different predatory situations: Discordant (as seen in Figure 1), Control (as seen in Figure 2), and Concordant (as seen in Figure 3). The hermit crabs were placed into the testing chamber, and once they were out of their shells for 30 seconds straight, the stimuli video would begin playing. Once the video had been playing for 20 seconds, the crab’s shell would be lightly tapped with a glass rod until it hid in its shell. The time from the tap until it came back out of its shell was then recorded in the data table below. This stimulus timeline is summarized visually in Figure 4 wherein the same trial timeline was used as Ryan et. al, except that the tapping method was added.

In their study of the same species (Coenobita clypeatus), Ryan et. al, found that with the same predatory cues presented, the mean latency to respond to the stimuli was lowest for the Concordant trials, followed by their Control trial, and then the Discordant trial (2012). This study was conducted to model the relationship of how the spatial distance between distractor and target affect hermit crab anti-predator response. Therefore it was hypothesized that, if we vary the direction between an auditory and visual stimulus, Hermit Crabs will exhibit different times of latency before freezing in response to the stimulus, with the fastest response time being the concordant trial. This response is due to anti-predator withdrawal behavior exhibited in hermit crabs creating a slowed response.

In the preliminary study, I found a c-clamp to be an inadequate way of holding the crabs in place during testing. Specifically during data collection, the crabs would try to escape instead of displaying the expected response of hiding in their shell when exposed to predator cues. Given the limitations of the c-clamp, it was clear that I needed to adopt a new model for the testing area. Wale et. al’s study on foraging behavior in the shore crab (Carcinus maenas) was then adapted to replace the c-clamp for the final study (as seen in Figures 1,2 and 3).

Figure 1 represents the experimental set-up for the Discordant trials in which an iPad showing the visual stimuli is set up on the left, longer side of the tank. A speaker playing the audio white noise stimulus is placed behind the mesh screen on the shorter side of the tank. It is important to note that in the Discordant trial, the audio and visual stimuli are coming from different directions. The Control Trial in Figure 2 only has the visual stimulus against the left side of the tank. Lastly, the Concordant trial in Figure 3 has the same experimental set up as Figure 1, except that the visual stimuli is moved to be against the mesh screen so that the visual and auditory stimuli are coming from the same directions. This experimental set up and trials were used from Ryan et. al.

As seen in Figure 4, the median time for a hermit crab to re-exit its shell from lowest to highest was Discordant, Control, and Concordant. This was the opposite of what was expected based on Ryan et. al’s experiment (2012). It was hypothesized that this unexpected outcome was caused by the order in which each trial was conducted, as Concordant was done first, then Discordant, and then Control for all test subjects. To further test this hypothesis, twelve more trials were conducted for four of the nine subjects where the testing order was changed and rotated for each individual. After a break of 3 weeks, the four crabs were exposed to the Control stimuli and then either the Concordant or Discordant stimuli to see if the order influenced the time it took to re-emerge from the shell. In order to determine if the trial order changed the results, two tailed independent T-tests were conducted to compare re-exit times between the sequential data and the values for the mixed order retrials. It was found that there was no significant statistical difference for the re-exit times for the both the Discordant and Concordant stimuli vs Control (p > .05 for each respectively), however there was a significant difference between the Discordant and Concordant trials (p < .05) (as seen in Figure 6). Although a significant difference was not shown for all of the trials, the difference in the Concordant and Discordant Trials provides an opportunity for future research into whether Hermit Crabs are able to learn after multiple exposures. Although this was just a small test set of subjects, it appears as though the p-values changed significantly between the sequential trials and the mixed retest trials. Previous studies have suggested that Hermit Crabs are able to learn behavior, such as Mark V. Tran’s study of Behavioral Interactions to Novel Food Odors (2015). However, the studies from which the experiment was based, did not discuss this finding as being relevant to their experiment. Thus, the order in which the trials were presented may have caused the hermits to learn what to expect from the predator stimuli, creating smaller reactions and thus shorter times for crabs to re-exit their shell since there were no real life repercussions for exiting the shell more quickly after being tapped. This discovery implies that Hermit Crab conditioning and learning may be a fruitful avenue for future research such as seeing if hermit crab shell size has an effect on this experiment.

Although this study created some significant results, it has limitations considering the fact that it was adapted to a school environment and schedule. The rest times between testing days were not consistent between all hermit crabs and trials, as it was on a high school schedule. If this experiment were to be done again, each hermit crab should have equal and consistent rest times between testing days or tested as a knowledge gap to see if it had a significant alteration to results. This research found that the spatial relationship of an auditory distractor to the visual predator affected hermit crab’s response time to re-exit their shells. This is a particularly exciting discovery because it showed that resembling spatial cueing and attentional capture can be demonstrated in an invertebrate. This study expands our understanding of how components (e.g., duration, loudness) of extraneous acoustic stimuli predictably impact anti-predator behavior, a finding with both theoretical and applied value. Overall, this research is increasingly significant as anthropogenic noise has increased as technology has advanced, reducing the fitness of the hermit crab species, possibly leading to extinction at the hands of prey.

SUPPLEMENTAL INFO AND FIGURES:

Link to Lab Notebook

https://docs.google.com/document/d/1L7OlQO35U11Zgeu1CW4h-WtWcqjdYe-A54fDzv6OYYw/edit?usp=sharing

WORKS CITED:

Works Cited

Billock, W., Dunbar, S., Shell and food acquisition behaviors: Evidence for Contextual Decision Hierarchies in hermit crabs, Journal of Experimental Marine Biology and Ecology, Volume 398, Issues 1–2, 2011, Pages 26-32,

Gorman, D., Ragagnin,M.N., McCarthy, I,D., Turra, A, (2018). Risk-taking and risk-avoiding behaviors by hermit crabs across multiple environmental contexts, Journal of Experimental Marine Biology and Ecology, Volume 506, Pages 25-29.

Hingham, T., Stewart, J., & Wainwright, P. (2015). Turbulence, Temperature, and Turbidity: The Echo Mechanics or Predator-Prey Interaction in Fishes, Integrative and Comparative Biology.

Kats, L.B., & Dill, L.M., (1998) The scent of death: Chemosensory assessment of predation risk by prey animals, Écoscience, 5:3, 361-394.

Magee, B & Elwood, R.W (2016). Trade-offs between predator avoidance and electric shock avoidance in hermit crabs demonstrate a non-reflexive response to noxious stimuli consistent with prediction of pain, Behavioural Processes, Volume 130, Pages 31-35, ISSN 0376-6357,

Putman, B., Coss, R., Clark, R. (2015) The ontogeny of antipredator behavior: age differences in California ground squirrels (Otospermophilus beecheyi) at multiple stages of rattlesnake encounters, Behavioral Ecology and Sociobiology.

Ping, X., Sun Lee, J., Garlick,D., Jiang,Z., Blaisdell, A.P (2015). Behavioral evidence illuminating the visual abilities of the terrestrial Caribbean hermit crab Coenobita clypeatus, Behavioural Processes, Volume 118, Pages 47-58, Dukas, R., Kamil, A.C., (2000). The cost of limited attention in blue jays, Behavioral Ecology, Volume 11, Issue 5, Pages 502–506.

Pereyra, P., González Portino, E., Maldonado, H., (2000). Long-Lasting and Context-Specific Freezing Preference Is Acquired after Spaced Repeated Presentations of a Danger Stimulus in the Crab Chasmagnathus, Neurobiology of Learning and Memory, Volume 74, Issue 2, Pages 119-134.

Ryan, K., Blumstein, D., Blaisdell, A & Stahlman, D (2012). Stimulus concordance and risk-assessment in hermit crabs (Coenobita clypeatus): Implications for attention, Behavioural Processes, Volume 91, Issue 1, Pages 26-29.

Rimmer, J., Todd, C., Shuker, D., (2021). Context-dependent use of visual cues in the shell selection behaviour of the hermit crab Pagurus bernhardus, Behavioural Processes, Volume 188.

Sagamiko, T., Kozlov, V., Wennergren, U.,(2021). Predator survival analysis of a Prey-Predator system with prey species pool, Scientific African, Volume 14.

Stahlman, D. W., Yim-Hol Chan, A., Blumstein, D.T, Fast, C.D & Blaisdell, A.P (2011). Auditory stimulation dishabituates anti-predator escape behavior in hermit crabs (Coenobita clypeatus), Behavioural Processes, Volume 88, Issue 1, Pages 7-11.

Shettleworth, S., (2002). Animal cognition and animal behaviour, Animal Behaviour, Volume 61, Issue 2, Pages 277-286.

Shragai, T., Ping, X., Arakaki, C., Garlick, D., Blumstein, D. T., & Blaisdell, A. P. (2017). Hermit crab response to a visual threat is sensitive to looming cues. PeerJ.

Tidau, S & Briffa, M (2019). Anthropogenic noise pollution reverses grouping behaviour in hermit crabs, Animal Behaviour, Volume 151, Pages 113-120.

Tidau, S., Briffa, M., (2019). Anthropogenic noise affects risk assessment and attention: the distracted prey hypothesis Biol. Lett.6458–461.

Tomsic, D., Sztaker, J., Berton de Astrada, M., Oliva, D., Lanza, E. (2017). The predator and prey behaviors of crabs: from ecology to neural adaptations. The Company of Biologists.

Ydenberg RC, Dill LM, 1986. The economics of fleeing from predators. Adv Study Behav 16: 229-249.

Yim-Hol Chan, A , Stahlman, D.W., Garlick, D., Fast, C.D., Blumstein, D.T. & Blaisdell, A.P (2010). Increased amplitude and duration of acoustic stimuli enhance distraction, Animal Behaviour, Volume 80, Issue 6, Pages 1075-1079.

Yim-Hol Chan, A., Giraldo-Perez, P., Smith, S., Blumstein & D.T (2010). Anthropogenic noise affects risk assessment and attention: the distracted prey hypothesis, Biology Letters, Volume 6, Issue 4.