PRESENTATION

BACKGROUND

Social learning, the act of learning a behavior by observing others, is a crucial part of childrens’ development, allowing them to learn behaviors such as aggression, cooperation, social interaction (Editors of Encyclopedia Britannica, 2019). This mechanism of learning behaviors has many positive outcomes. For example, a study by Bandura et al. found that children who witnessed another individual being punished for aggressive behavior were less likely to act aggressively than children who saw rewards for such behavior (Editors of Encyclopedia Britannica, 2019). However, more research is necessary to quantify the extent to which this is true. This indicates that children tend to avoid behaviors that they see receive a negative response, and copy behaviors that receive a positive one. However, social learning is not limited to just children. After analyzing previous research around how executives grow and change over the course of their careers, in 2022 the Leading Effectively staff at the Center for Creative Leadership created the 70:20:10 model for workplace learning, which indicates that social learning makes up 20% of learning in the workplace (Leading Effectively Staff, 2022). Specific results of social learning in the workplace include promoting positive work environments, increased productivity, and increased knowledge retention (UC Berkeley, 2023). Clearly, social learning is a widespread phenomenon, utilized by people of all ages in a variety of environments. Understanding this phenomenon will improve the fields of psychology, education, and social work by providing insight into an effective behavior and learning method.

Because of the important role social learning plays in human development and behavior, it is important to understand what environmental and psychological factors encourage a person to copy someone else’s behavior. A greater understanding of the mechanisms behind social learning and how reinforcement over time affects the permanence of learned behaviors can provide new insights into negative behaviors such as bullying, aggression, violence, and crime (Leff et al. 2009). All of these behaviors are theorized to be a result of social learning by researchers such as Alburt Bandura in 1963 and Ruth Triplett in 2015 who investigated social learning’s effect on aggression and crime, respectively (EdX, 2021; Triplett, 2015). In his study, Bandura exposed young children to adults acting aggressively towards a doll, and found that those children replicated that behavior (EdX, 2021). In her study, Triplett found that individuals who commit crimes did so because they learned that behavior through others (Triplett, 2015). This work suggests that increased knowledge of what causes and encourages the social learning of a new behavior can teach us how to prevent learning of negative behaviors early on (Western Governors University, 2020; EdX, 2021). Social learning can also be used to enforce positive behaviors. For example, social workers at EdX proposed that spending time around role models who advocate for social change can inspire individuals to do the same (EdX, 2022). However, before we can actually utilize social learning as a tool to better society, it needs to be studied further, as we do not yet understand how reinforcement over time impacts social learning (Leff et al. 2009). Unfortunately, due to ethical concerns and their behavioral variability, humans are not the easiest test subjects for well controlled behavioral studies (Sterelny, 2019). Thus, a model organism is required.

Zebrafish (Danio rerio) are particularly good model organisms for studies surrounding neuroscience, behavior, and, importantly, social learning, as they have many similarities to humans. Humans have significant genetic similarity to zebrafish, as over 70 percent of human genes contain a zebrafish orthologue, and humans and zebrafish are proposed to have some corresponding brain structures (Meshalkina et al., 2017; Adams et al., 2018). Like mammals, zebrafish also have an integrated nervous system, which carries out the processing of sensory information, including visual cues, making them effective model organisms for studies involving visual stimuli (Adams et al., 2018; UMass Dartmouth, n.d.). Humans and zebrafish also have many behavioral similarities. For example, both zebrafish and humans are very social, which makes them effective model organisms for studying social interactions (Meshalkina et al., 2017). In zebrafish, this sociableness is observable in their shoals, groups of zebrafish that live together and interbreed. The size of these groups are highly variable and can range from a few to a few hundred individuals, which illustrates the complexity of their social interactions (Ghoshal et al., 2021). Additionally, when placed in new or unfamiliar environments, zebrafish often exhibit fear and stress, emotions that are very common in humans (Meshalkina et al., 2017). For example, if a zebrafish is stressed, it tends to stay towards the bottom or the walls of the tank, to stick other members of its shoal, or, if extremely stressed or afraid, it may freeze randomly (Slabbekoorn, 2020). Lastly, like humans, zebrafish can actually have distinct temperaments. Personality traits such as boldness, their willingness to take risk or desire to explore surroundings, vary between individual fish (Slabbekoorn, 2020; Boldness Definition, n.d). To investigate these traits further, researchers in a 2020 study on zebrafish stress physiology, created an experimental tank with speakers to stimulate a stressful environment using noise pollution. They placed the fish in the tank to analyze their reactions to the noise and measure the unique effects of a stressful environment on fish behavior. They observed a variation in reactions to the noise pollution, which confirms that zebrafish have distinct temperaments that reflect a cognitive complexity not normally associated with such simple organisms (Slabbekoorn, 2020).

In order to use a model organism for a behavioral study, it is important to have a sufficient understanding of that organism’s cognition, which is how they acquire, manipulate, store, and retrieve information (Filosa, 2016). A 2021 study by Santacà et al. helped expand our knowledge of zebrafish cognition by researching zebrafish’s visual discrimination learning. The researchers used an appetitive conditioning paradigm, a model to demonstrate the association between a reaction to a stimulus and a reward, to teach zebrafish to discriminate between outlined shapes of different colors (Santacà et al., 2021; Martin-Soleach et al., 2007). The experimental setup of this study consisted of an experimental tank with two stimuli on opposite sides: a circle with a red outline and a circle with a green outline (Santacà et al., 2021). To train a red zebrafish, the researchers placed it individually in the experimental tank and allowed it to explore. When the fish approached (i.e. within the distance of one body length) the red circle, they were rewarded with one drop of live brine shrimp larvae (Artemia salina). The red fish were not rewarded if they approached the green circle, and were exclusively trained to select for the red circle, which was repeated every day for twelve days. This procedure was replicated for the green fish with the green circle, respectively (Santacà et al., 2021). The researchers found that the trained zebrafish had a 67.79 ± 3.65 % accuracy in color discrimination of outlined shapes, and their accuracy improved over the course of the experiment (Santacà et al., 2021). This study verified that zebrafish are capable of distinguishing between various colors of outlined shapes and can be trained to select for certain colors based (Santacà et al., 2021). 

Santacà et al.’s study provides ample information about zebrafish cognition and it clearly demonstrates that zebrafish are an effective model organism for studies testing visual discrimination learning. Because they are so well suited to studies of this nature, there will undoubtedly be many future studies in which researchers train zebrafish to discern two colors, objects, et cetera. In their study regarding visual discrimination, Colwill et al. point out that it is important to understand which reinforcement strategies zebrafish best respond to in order to reduce training time, which is a complicated and time consuming endeavor (Colwill et al., 2005). Discovering more effective ways to train zebrafish is critical to make research easier and more efficient, as they are becoming an increasingly popular model organism for studies on behavior, learning, and memory (Kit Tan et al., 2022).

Unsurprisingly, considering their cognitive and behavioral similarities to humans, zebrafish also use social learning as a survival mechanism. It allows them to gather more data about food, predators, and the environment by observing their companions’ behavior and copying it if it proves successful (Zala et al., 2012). Additionally, social learning provides a selective advantage as learning through observation allows zebrafish to avoid the costliness of asocial learning, which is when learning occurs through trial and error (Zala et al., 2012). One study conducted by Zala et al. in 2012 examined different social learning strategies in wild and domesticated zebrafish (Zala et al., 2012). The researchers investigated whether exposure to bolder domesticated zebrafish resulted in wild zebrafish becoming bolder when approaching a moving object. For the controls, a group of wild zebrafish and a group of domesticated zebrafish were placed in separate tanks, and the researchers manually waved a circular blue object in front of each tank twenty times (Zala et al., 2012). The less bold wild fish kept an average of 28±2 cm away from the object, which was significantly farther than the bolder domesticated fish, who kept an average of 17±6 cm away from the object. These controls established that the wild zebrafish were less bold when approaching a novel stimuli (Zala et al., 2012). For the experimental arm, the wild and domesticated fish were placed in a tank in groups of four and two, respectively, and the researchers distinguished the two groups by placing orange tags on the domesticated fish. As with the controls, a circular blue object was waved manually in front of the tank twenty times and the approach distance was measured (Zala et al., 2012). In accordance with the researchers hypothesis, after witnessing the bold behavior of the domesticated fish, wild fish socially learned bolder behavior and decreased their distance from the novel object, averaging 25±3 cm (Zala et al., 2012). Though a short distance, this was found to be statistically significant (Zala et al., 2012). One knowledge gap that persists in this study is that it is unclear how long these behavioral changes last. The researchers could address this question by testing the wild zebrafish everyday for a week as opposed to only once after shoaling with the bolder domesticated fish. By doing this, they would be able to see how well the zebrafish could remember and replicate the behavior they witnessed over longer periods of time. More specifically, it is important to know whether these behaviors are only adopted temporarily, or if social learning leads to long term behavioral adaptations. These findings would be crucial and transferable to social learning’s application to humans, especially if it can be leveraged to teach and enforce positive behaviors. By gaining a better understanding of the permanence of behaviors learned socially, we can alter social learning techniques to make them more effective long term. 

But before that research can be conducted, it was first necessary to learn more about social learning in zebrafish. Herein I will describe a replication of Lindeyer et al.’s study, which tested the social learning of escape routes in zebrafish (Lindeyer et al., 2010). The hypothesis was that naïve zebrafish would follow and remember the escape route demonstrated to them by a trained zebrafish to escape a novel predator with more success than an untrained fish. This is due to the fact that zebrafish within a shoal exhibit behavioral homogeneity to minimize predation risk (Lindeyer et al., 2010). This study contained elements from both Santacà et al.’s 2021 study regarding visual discrimination and Zala et al.’s 2012 study, which tested social learning in zebrafish. As with Santacà et al.’s study, the replicate study began by training two groups of fish to escape a novel predator (a trawl net) through different colored escape routes: red and yellow. To demonstrate mastery, zebrafish had to consistently escape through their respective route in under thirty seconds (Lindeyer et al., 2010). Trained demonstrators were then placed in the experimental tank with untrained test subjects, and due to social learning, these test fish followed the demonstrators through the correct escape route (Lindeyer et al., 2010). Next, they removed the demonstrator fish and tested the subjects on their own (Lindeyer et al., 2010). The hypothesis was supported. The test subjects remembered the escape route shown to them by a demonstrator fish, whereas negative control test fish who witnessed untrained demonstrators escaped at random. 

This replication confirms what previous studies had already established: zebrafish can acquire a behavior through social learning. However, it is important to research how long these behaviors last. In order to test zebrafish’s long term memory of socially learned behaviors, one would replicate Lindeyer et al.’s study, modified to only one group of trained demonstrator fish for simplicity. After observing and following the demonstrator fish’s successful escape from the novel predator, the test fish would then be tested on their own on days 0, 1, 3, and 7. The hypothesis is that these test fish will remember the behavior that allowed them to escape from a life threatening situation and thus will continue escaping through the correct escape route for the duration of the experiment.

Due to the cognitive and behavioral similarities to humans, zebrafish are ideal model organisms for studying social learning, which has crucial applications to the field of psychology (Adams et al., 2018; Meshalkina et al., 2017; Kauleff et al., 2013; Gerlai et al., 2023). By leveraging zebrafish models, researchers can gain a better understanding of the underlying causes of this behavior, its success as a survival mechanism, impacts on group behavior, and most importantly, lasting effects on individuals (Zala et al., 2012). These findings can be applied to humans, as children use social learning to develop certain behaviors, such as cooperation and sharing, and adults utilize social learning to increase productivity (Editors of Encyclopedia Britannica, 2019; UC Berkeley, 2023). Additionally, researchers have proposed that social learning can prevent negative behaviors, like aggression and crime, and promote positive ones, like advocating for change, by showing individuals the consequences of participating in these activities (EdX, 2021; EdX, 2022). As such, a complete understanding of the association between long term memory and socially learned behavior is critical. It would allow us to alter social learning techniques to ensure lasting positive effects, which could have a large impact on individuals, school and work communities, and societies as a whole. Thus, this research in progress aims to fill a crucial knowledge gap on this widespread and innate behavior so that we can learn how to use it to our advantage.

METHODS & PROCEDURE

Safety Protocol

While conducting this experiment, gloves and goggles are to be worn at all times. Gloves are used to prevent irritation of the skin. Because the procedure involves glassware, it is also important to wear goggles to protect the eyes should any of the tanks or beakers break or shatter.

Overview

Preliminary research conducted in 11th grade replicated Lindeyer et al.’s 2010 study testing social learning habits in zebrafish (Danio rerio). This preliminary replicate study investigated whether untrained zebrafish would continue exhibiting a behavior shown to them by a trained demonstrator without that demonstrator present. This research question was addressed by allowing test fish to follow a trained demonstrator through an escape route of a certain color (red or yellow). After five minutes, the zebrafish were tested to see if they continued escaping through the correct route without the demonstrators present. Research conducted in 12th grade built off of those initial findings to examine whether test fish who followed a trained demonstrator will remember the correct escape route over longer periods of time. Herein it was hypothesized that the zebrafish will continue escaping through the correct escape route because that behavior allowed them to avoid predation. To investigate this hypothesis, test fish were assessed one hour and one week after following the trained demonstrator. 

Tank Set Up

All fish tanks were purchased from Carolina Biological. The 20 gallon tank (Item # 671234) was used to acclimate the fish, and was referred to as the set up tank. The two 2.5 gallon tanks (Item # 671226) were used as the Red and Sham Demonstrator home tanks. The three 10 gallon tanks (Item # 671230) were used as the Red Test Subjects home tank, Sham Test Subjects home tank, and the testing aquarium. Size of the testing aquarium for both the replicate and novel study was decreased from the testing aquarium used in Lindeyer et al.’s 2010 study (150 x 50 x 30 cm) due to space constraints in the classroom setting. The replicate study used a 30 x 21 x 15 cm (2.5 gallon) testing aquarium. The limited space in the tank led to a significant decrease in escape time, as the fish were closer to the escape routes. In Lindeyer et al.’s study, the average escape latency of test fish with trained demonstrators was 69.4± 6.3 seconds (2010). Whereas in the replicate study, their average escape latency was 7± 4.5 sec. In order to more closely mimic the conditions of Lindeyer et al.’s study (2010), the novel study used a 41 x 25.5 x 30 cm (10 gallon) fish tank for the testing aquarium. The larger tank size was selected to increase the escape time for both demonstrator and test fish in all arms during the novel study. Notably, longer average escape times are less affected by random fluctuations in fish behavior, which will decrease the percent error associated with these  measurements. This provides researchers greater opportunity to accurately measure significant differences across arms.

Houseran wire cutters (ASIN B0BHKRK4C4) were used to cut 1 cm slits on the front and back walls of the testing aquarium to insert dividers into during training and testing. The slits were made 20 cm from the right and left walls of the testing aquarium. The right side of the testing aquarium received a SunGrow plastic plant (ASIN B01MUBOZ3X) and a Boxtech breeder box (ASIN B07FXTB65Z) used to house a model shoal to act as stimuli during training and testing (see Figure 1 below).

All tanks were filled with conditioned tap water made with 1055 uL each of API Water Conditioner (ASIN B07BTZQ1Z2), API Ammo Lock (ASIN B000255N0A), and Seachem Tank Stabilizer (ASIN B01HHDCT1A) per 2.1 gallons of water. Measurements were taken using a 1000 μL micropipette. All tanks received a Hygger heater (ASIN B07X45XC3Z) set to 26.5±0.5 °C (Vergauwen et al., 2010), the set up tank received a Marineland filter (ASIN B07YXJ1WLH), the home tanks and testing aquariums received a Datoo filter (ASIN B091DQTJYC) and the set up and home tanks received a Tetra Whisper air pump (ASIN B0009YF4FI) (Avdesh et al., 2012; Hariharan et al., 2024).

Aquarium Divider Construction

To construct the testing aquarium dividers, a design was created on Adobe Illustrator (https://www.adobe.com/products/illustrator.html) on an Apple desktop. Adobe Illustrator was utilized because it allowed researchers to precisely scale designs for the dividers on the desktop. The design for the training divider consisted of a 23.5 x 32 cm rectangle with two 8 x 18 cm rectangles within it, 14 cm from the top, 2.6 cm from either side, and 2.6 cm apart. The design for the testing divider consisted of a 23.5 x 32 cm rectangle with two 4.75 x 7 cm rectangles within it, 25 cm from the top, 4.6 cm from either side, and 4.6 cm apart. A KliHDSM 24 x 48 in white plexiglass sheet (ASIN B00IW9691K) was sawed in half using the Walter Meier bandsaw (Item #JWBS-14DXPRO). The bandsaw is a dangerous tool due to its sharp blade. To prevent injury, researchers passed the BERT Bandsaw Training Test (https://bert.thebetalab.org/) and wore safety goggles.

One half of the acrylic sheet was placed in the Glowforge Pro laser cutter (https://shop.glowforge.com/products/glowforge-pro) and the design from Adobe Illustrator was exported to the laser cutter. For both the training and testing dividers, the left holes were outlined with red Scotch electrical tape, and the right holes with yellow Scotch electrical tape (ASIN B001B19FDK).

To create the clear escape route blocks for training and testing, clear Polystyrene plastic (Mega Format; ASIN B0968ZFFBZ) was cut into two 22.5 x 27 cm pieces using the Walter Meier bandsaw (Item #JWBS-14DXPRO).

During training and testing, the divider and escape route blocks were inserted into the slits in the rim of the testing aquarium, such that the blocks covered the escape routes. To prevent fish from swimming through the gap between the left and right edges of the aquarium divider and walls of the testing aquarium, two 25 cm pieces of 5/15” Flinn plastic tubing (Item #AP5426) were taped to the edges of the divider using Gorilla Tape (ASIN B07LFRN1K8).

Zebrafish Acclimation and Care

Adult zebrafish (Danio rerio, https://www.wetspottropicalfish.com/product/danio-rerio-2/) were purchased from the Wet Spot. Upon arrival, the fishing bag was floated in the set up tank to warm up (Zebrafish International Resource Center). After 10 minutes, zebrafish were switched from their shipping bag to a 2000 mL beaker, along with the shipping water. Every 10 minutes for 40 minutes, 200 mL of their shipping water was replaced by 200 mL of tank water from the set up tank. The demonstrator fish were removed from the 2000 mL beaker and placed in the set up tank using a trawl net. After one week, the demonstrator fish were transferred from the set up tank to their respective home tanks, (i.e. four demonstrator fish to the Red Demonstrators Home Tank and four demonstrator fish to the Sham Demonstrators Home Tank). The same process was repeated with the orange test zebrafish (Danio kyathit; the Wet Spot; https://www.wetspottropicalfish.com/product/danio-sp-kyathit/) but with the Red and Sham Test Subjects Home Tanks, respectively. 

To ensure the health and safety of the zebrafish, all tanks were checked on once per day to ensure that no fish had died. If a fish was found dead, it was removed immediately from the tank and discarded in the trash. Once every two days, all fish were fed one pinch of Tetra Tropical Flakes (Item # 77101), as recommended by Boisen et al. (2003). Once every week, ~30% of the tank water was replaced with freshly made conditioned water (Washington University, 2022). Once every two weeks, water was tested for heightened ammonia levels using an API ammonia test strip (ASIN B002DREWMK). If ammonia levels were high, water changes were conducted daily for three days (Washington University, 2022).

Training

For initial training in the replicate study, the testing aquarium was divided by the divider with enlarged escape routes, as suggested by Lindeyer et al. (2010). Use of enlarged escape routes was omitted for the novel study, as the size of the enlarged escape route (8 x 18 cm) was too large in comparison to the size of the fish (~2.5 cm). Qualitatively, fish were observed to swim through the large escape routes without choosing, as opposed to intentionally selecting the route with the correct color.  This was most often measured by extremely variable and short escape times prior to any significant training to justify their behavior. Thus, the large escape routes were designated ineffective and were a confounding variable that was eliminated by utilizing the divider with smaller escape routes (4.75 x 7 cm) more close in size to the subjects tested.

All training sessions were filmed on an iPhone. Each training session consisted of twelve two minute rounds and occurred twice per day. This was increased from Lindeyer et al. ’s recommended four two minute rounds in order to maximize training time and efficiency (2010). Longer training sessions had no negative impacts on the fish’s performance or energy, but they were shown to have significant improvements on their learning. At the conclusion of each training round, zebrafish were ushered back to the left side through the red escape route using the trawl net. This was modified from Lindeyer et al.’s 2010 study, which lifted the partition before ushering the fish back to the left side, to reinforce the behavior of swimming through the red escape route. Between training rounds, fish were given thirty seconds to acclimate. During this time, both escape routes were covered with clear plastic barriers. During the two minute training round, the plastic barrier covering the red escape route was removed and a trawl net (i.e. the novel predator) moved back and forth from the left wall to the divider at a constant pace of one second (i.e. one second from left wall to divider, one second from divider to left wall). This trawl net technique was modified slightly from Lindeyer et al.’s study, which recommended moving the trawl net at a constant pace of 15 seconds with 15 second pauses at the left wall and divider (i.e. 15 seconds from left wall to divider, 15 second pause, 15 seconds from divider to left wall, 15 second pause) (2010). This was a necessary modification, as when the 15 second timing was used during training, the demonstrator fish exhibited no fear of the slow-moving net and had no desire to escape from it. Increasing the speed from the published approach successfully alarmed theis fish, and prompted it to figure out a means of escape. This training method was a form of associative learning: the learning of a process or behavior (e.g. swimming through an escape route) due to another element (e.g. the danger of a novel predator) (Shimada et al., 2022).

If at any point during the procedure, the zebrafish struggled to escape (i.e. it took them longer than 10 seconds to escape), they were directed towards the red escape route with the trawl net. This was also modified from Lindeyer et al.’s 2010 procedure, which suggested directing the fish towards the red escape route using a paintbrush, as the fish were easily able to avoid the small brush.

Once fish demonstrated mastery of escape (i.e. escaping in under ten seconds through the correct route for two training sessions), the final red demonstrator selection process could begin.

Red Demonstrator Selection Process

To determine which Red Demonstrators had the strongest understanding of the escape route, each fish was tested individually. The testing procedure was identical to the training procedure, except that each fish’s tested session consisted of three rounds, not twelve, due to timing constraints. At the conclusion of each testing session, each fish’s average escape time was calculated using the data from the videos. Any fish with an average escape time of greater than 10 seconds was disqualified from being used for the positive control. The first two fish with an average escape time of less than 10 seconds were selected as the official Red Demonstrators.

For the replicate study, researchers used both red and yellow demonstrators. The corresponding test fish yielded similar results during experimental arms. Red Test Subjects had an escape latency of 17±26 seconds and a red:yellow escape route preference of 4:0, which was statistically significant (p<0.05). One Red Test Subject behaved significantly differently to the other three test subjects during one trial, which explains the high standard deviation and warrants analysis for outliers. With that data point omitted, their average escape latency was 8±8 seconds. Yellow Test Subjects had an escape latency of 4±2 seconds and a yellow:red escape route preference of 3:1, which was statistically significant (p<0.05). As a result, the Yellow Demonstrator and Test groups were omitted from the novel study due to time constraints.

Positive Control

For the positive control of this research study, four naive test fish (Red Test Subjects) were tested with two trained demonstrators (Red Demonstrators). The purpose of this arm was to establish that untrained test fish will copy the behavior of trained demonstrator fish if that behavior allowed the demonstrators to escape predation (Lindeyer et al., 2010; Roy et al., 2017). Because they witnessed the Red Demonstrators escape from the novel predator through the red escape route, the Red Test Subjects should copy their behavior, and thus have a similar escape time and escape route preference to the trained demonstrators. Using four test subjects allows for sufficient data to be collected should there be any confounding variables with a singular fish. Additionally, Lindeyer et al. found that witnessing two demonstrators led to the highest success among test fish (2010). 

The experimental tank was prepped with the testing aquarium divider and clear barriers covering both escape routes. Two Red Demonstrators and four Red Test Subjects were placed on the left side of the testing aquarium and were given two minutes to acclimate. A one minute stopwatch was started, the clear barriers were removed, and a trawl net was placed at the leftmost end of the tank. The trawl net moved back and forth from the leftmost side to the barrier over the course of one minute (~2 seconds per period). This was shortened from the two minute trial time in the replicate study due to timing constraints. However, it should not have any effect on experimental results, as all fish who escaped in the positive control in the replicate study escaped in under one minute. At the end of the one minute timer (or, once all fish had escaped), the clear barriers were put back, and all fish were transferred back to the left side of the tank using the trawl net. The fish were given one minute to acclimate, and the process was repeated eight times. At the conclusion of testing, all fish were placed back in their respective home tanks. During the experimental arms, these test fish should continue replicating the trained demonstrators’ behavior because that behavior allowed them to escape predation.

All trials were filmed on an iPhone. All data (i.e. escape time, escape route chosen, and number of fish escaped) was recorded in a data table. Videos were rewatched as necessary to complete the data tables.

Negative Control

For the negative control of this research study, four naive test fish (Sham Test Subjects) were tested with two untrained Sham Demonstrators. The independent variable (i.e. the trained demonstrators) was eliminated to establish baseline behavioral results for test fish in the testing aquarium with the moving trawl net (Lindeyer et al., 2010; Zala et al., 2012). Thus, there should be no difference in escape time or escape route preference between the Sham Test Subjects and Sham Demonstrators. However, Sham Test Subjects should have a longer escape time and more variable escape route preference than Red Test Subjects.

The negative control of this research study had an identical procedure to the positive control, except with Sham Demonstrators and Sham Test Fish, respectfully. The decreased trial time may impact negative control results, as in the negative control of the replicate study, Sham Test Fish escape times ranged from 4 to 120 seconds.

During the experimental arms, the Sham Test Fish should continue to escape at random since they have not learned any other behavior.

Experimental

The first and only experimental arm of Lindeyer et al.’s study and the replicate study investigated whether the test subjects could retain the behavior acquired socially (i.e. the escape through the correct escape route) after five minutes without the demonstrators present. Due to timing restraints, this experimental arm could not be conducted during the novel study.

The first experimental arm of the novel study investigates whether the test subjects could retain the behavior acquired socially (i.e. the escape through the red escape route) after one hour without the demonstrators present. The same procedure from the negative and positive controls was repeated with the Red Test Subjects without the Red Demonstrators. This process was then repeated with the Sham Test Subjects without the Sham Demonstrators.

The second experimental arm of this study tests whether the test subjects could retain the behavior acquired socially (i.e. the escape through the red escape route) after one week without the demonstrators present. The same procedure from the first experimental arm was repeated with the Red Test Subjects and Sham Test Subjects.

Data analysis

At the conclusion of testing, all data collected was imported to Google Sheets. The two variables being analyzed were escape route preference and escape latency (Lindeyer et al., 2010). Escape route preference was defined as the number of fish who escaped through each route per trial. Escape latency was defined as the time from when the trawl net entered the tank to the time when the last fish escaped, in seconds. If a singular fish has a drastically different escape time (i.e. >45 seconds) then all other fish in that cohort during a singular trial, then it can be constituted as an outlier and omitted from data analysis. The average and standard deviations for escape latency and escape route preference were calculated for each group per arm. 

For the negative control, a two tailed correlated t test was used to compare the Sham Test Subjects’ preference of red over yellow. As the negative control measured a baseline result for untrained fish, the p value should be statistically insignificant (p>0.05) showing no preference for either escape route. For the positive control, a two tailed correlated t test was used to compare the red test subjects’ preference of red over yellow. These results should be statistically significant (p<0.05), as the positive control is meant to show that test fish will follow trained demonstrators. The experimental arms also used a two tailed correlated t test to compare the Red and Sham Test Subjects’ preference of red over yellow. According to the hypothesis, the Red Test Subjects' results should be statistically significant (p<0.05) for both experimental arms, as they should retain the behavior they learned from the trained demonstrator to avoid predation. Meanwhile, the Sham Test Subjects never learned to associate a specific route with predator avoidance and should therefore continue escaping at random. As follows, these results should be statistically insignificant (p>0.05) for both experimental arms.

A two tailed independent t test was used to compare the average escape latency of Red Test Subjects during the positive control with Sham Test Subjects during the negative control (Lindeyer et al., 2010). Because Red Test Subjects were following trained demonstrators who escaped quickly and Sham Test Subjects did not have a trained demonstrator to follow, these results should be statistically significant (p<0.05). A two tailed independent t test was also used to compare the average escape latency of Red Test Subjects during first and second experimental arms with Sham Test Subjects during first and second experimental arms, respectively. According to the hypothesis, the Red Test Subjects should continue escaping at the faster rate shown to them by the trained demonstrators because that behavior allowed them to avoid predation. The Sham Test Subjects should continue escaping at a slower rate because they never witnessed a predator-avoidance behavior to follow. Therefore, these results should be statistically significant (p<0.05).

RESULTS AND DISCUSSION

Preliminary research conducted in 11th grade sought to confirm Lindeyer et al.’s findings regarding social learning habits in zebrafish (2010). Researchers tested Red Test Subjects, Yellow Test Subjects, and Sham Test Subjects five minutes after exposure to their respective demonstrators on their escape route preference and escape latency.

The escape route preference of Red Test Subjects and Sham Test Subjects were analyzed using a Chi Squared Test. This method of calculating route preference differs from that used by Lindeyer et al. (2010), who measured route preference on a scale of -1 (all four Test Subjects escaped through red) to 1 (all four Test Subjects escaped through yellow) (2010). Because the Chi Squared test statistic (X2 = 5.556) was greater than the table value (X2 = 3.841), there is a significant preference for red over yellow for the Test Subjects who followed Red Demonstrators (Sinauer Associates, 2013). During the positive control, Test Subjects socially learned the escape route shown to them by the Red Demonstrators. This arm effectively established that untrained test fish will copy the behavior of trained demonstrator fish if that behavior allowed the demonstrators to escape predation (Roy et al., 2017). Moreover, it successfully replicated the research conducted by Lindeyer et al. (2010).

Meanwhile, during the negative control, Test Subjects witnessed no consistent behavior by the Sham Demonstrators to follow. As follows, the test statistic (X2 = 2.758) was less than the table value (X2 = 3.841) because there was no significant preference for either escape route (Sinauer Associates, 2013). This finding is logical as the Sham Demonstrators were untrained and chose each exit with near equal probability.This arm effectively established baseline behavioral results for test fish in the testing aquarium with the moving trawl net (Lindeyer et al., 2010; Zala et al., 2012)

The escape latencies of Red Test Subjects and Sham Test Subjects during the positive and negative control were compared using a two-tailed correlated t-test. Because the p-value (p = 0.00004) was less than 0.05, there was a significant difference between the two escape latencies. Thus, the positive and negative controls produced their expected outcomes. Red Test Subjects followed the behavior of Red Demonstrators during the positive control and escaped quickly, while Sham Test Subjects learned no behavior from the Sham Demonstrators during the negative control and escaped at a slower rate.

As shown by Figure 3A, Test Subjects exposed to Red Demonstrators continued exhibiting a strong preference for the red escape route (X2 = 25) five minutes after their first social learning experiment with the Red Demonstrators. Yellow Test Subjects showed statistically similar results to Red Test Subjects, and thus the corresponding data analysis has been omitted from these Results and Discussion, but can be found in the Supplemental Information section. Meanwhile, as shown by Figure 4A, Test Subjects exposed to Sham Demonstrators showed no preference of one escape route over another (X2 = 1.625). 

The escape latencies of Red Test Subjects and Sham Test Subjects five minutes after their respective experimental trials were compared using a two-tailed correlated t-test. Because the p-value (p = 0.027) was less than 0.05, there was a significant difference between the two escape latencies. Thus, the five minute post exposure experimental trials produced their predicted outcomes: Red Test Subjects continued escaping quickly, as shown to them by the Red Demonstrators during the positive control, while Sham Test Subjects continued to escape at a slower rate because they learned no behavior from the Sham Demonstrators during the negative control. These results show that zebrafish are capable of retaining a short term memory of a learned, life-saving behavior shown to them by a trained demonstrator. This is because social learning is used a necessary and evolutionarily advantageous survival tactic in the wild (Lindeyer et al., 2010). 

Novel research conducted in twelfth grade aimed to find whether Test Subjects would continue escaping through the escape route shown to them by a trained demonstrator over longer periods of time (one hour and one week). It was hypothesized that Test Subjects would continue escaping through the correct escape route and retain this learned behavior over long periods of time because this behavior allowed them to avoid predation, and thus would not be easily forgotten.

For the novel study, the positive and negative control followed the same protocol as the positive and negative control of the replicate study using new Test Subjects and the larger testing aquarium. These arms produced similar results to those the replicate study, accounting for slightly longer escape times due to the larger testing aquarium, thus they have been omitted from these Results and Discussion and can be found in the Supplemental Information section.

The escape route preference of Red Test Subjects and Sham Test Subjects one hour after exposure to their respective demonstrators were analyzed using a Chi Squared Test. Because the test statistic (X2 = 8.5) was greater than the table value (X2 = 3.841), there is a significant preference of red over yellow for the Test Subjects who followed Red Demonstrators during the positive control (Sinauer Associates, 2013). These Test Subjects socially learned the escape route shown to them by Red Demonstrators, and they were able to remember it after an hour because it was a life-saving behavior. Meanwhile, Test Subjects who followed Sham Demonstrators during the negative control witnessed no consistent behavior to follow. Thus, the test statistic (X2 = 1.6) was less than the table value (X2 = 3.841), showing there is no significant preference of either escape route for the Test Subjects who followed Sham Demonstrators during the negative control (Sinauer Associates, 2013). 

The escape route preference of Red Test Subjects and Sham Test Subjects one week after exposure to their respective demonstrators were analyzed using a Chi Squared Test. Because the test statistic (X2 = 15.385) was greater than the table value (X2 = 3.841), there is a significant preference of red over yellow for the Test Subjects who followed Red Demonstrators (Sinauer Associates, 2013). During the positive control, Test Subjects socially learned the escape route shown to them by Red demonstrators, and they were able to remember it a week later because it was a life-saving behavior. Meanwhile, during the negative control, Test Subjects witnessed no consistent behavior by the Sham Demonstrators to follow, as expected. Thus, the test statistic (X2 = 1.6) was less than the table value (X2 = 3.841), showing there is no significant preference of either escape route for the Test Subjects who followed Sham Demonstrators (Sinauer Associates, 2013). 

These results support the hypothesized outcome that Test Subjects will remember an escape route shown to them to escape a novel predator, and will continue escaping through that route over longer periods of time. Test Subjects who followed Red Demonstrators during the positive control showed significant preference of the red escape route during all experimental trials. In fact, during the novel study, Test Subjects’ preference of red over yellow actually increased over time, with the test statistic increasing from X2 = 8.5 to X 2 = 15.385, as shown in Figures 5A and 5B. Meanwhile, Test Subjects who followed Sham Demonstrators during the negative control showed no preference for either escape route during all experimental trials. As shown in Figures 2A, 4A, 6A, and 6B, all test statistics were less than the critical value (X2 = 3.841) (Sinauer Associates, 2013).

The escape latency results produced by the one hour post exposure tests in the novel study data did not support the hypothesis. Outliers from the data set (i.e. Test Subjects whose escape latencies lied outside of the 2nd and 3rd quartiles) were removed using an outlier test on Google Sheets. As a result, the Red Test Subjects’ one hour post exposure experimental only had five data points. Thus, a two-tailed independent t-test was used to determine whether there was a significant difference between the Red and Sham Test Subjects’ escape latencies. The p-value was 0.301, which is greater than 0.05, and therefore these results are not statistically significant, refuting the prediction that nine trials of prior exposure to trained demonstrators would help Red Test Subjects escape significantly faster than Sham Test Subjects.

One possible explanation for the unexpected escape latency results in the one hour post exposure experimental trials is that on the day these tests were conducted, Red Test Subjects were observed to be acting extremely lethargic, showing little fear or panic in response to the trawl net. Moreover, they had already undergone a round of testing an hour prior during the positive control. Due to timing constraints, researchers could not wait to conduct the tests on a day where Test Subjects appeared to be in better health and had to continue with the study.

The escape latency results produced by the data in the one week post-exposure tests in the novel study, however, supported the hypothesized outcome that Test Subjects who followed trained Red Demonstrators would escape more quickly than those who followed Sham Demonstrators. This may be because Red Test Subjects had time to rest and regain health, allowing them to escape at a faster rate. Outliers from the data set (i.e. Test Subjects whose escape latencies lied outside of the 2nd and 3rd quartiles) were removed using an outlier test on Google Sheets. As a result, the Red Test Subjects’ one week post exposure experimental only had five data points. Thus, a two-tailed independent t-test was used to determine whether there was a significant difference between the Red and Sham Test Subjects’ escape latencies. The p-value was 0.001, which is less than 0.05, showing there is a significant difference between the escape latencies of the two groups. This indicates that after one week, Red Test Subjects were able to remember and replicate the behavior shown to them by the Red Demonstrators, meanwhile the Sham Test Subjects escaped at a slower rate because they actually had to figure out how to escape by themselves.

There were two sources of systematic error in escape latency that must be addressed in these Results and Discussion. Firstly, for the replicate study, testing was conducted in a 2.5 gallon tank, while in the novel study, testing was conducted in a 10 gallon tank. This significant size increase explains an overall increase in escape latencies between the replicate and novel study. For example, Red Test Subjects had an average escape latency of 7 seconds during the positive control of the replicate study but 32 seconds during the positive control in the novel study. Secondly, during the replicate study, trials lasted 120 seconds, meaning Test Subjects could take up to 120 seconds to escape (Lindeyer et al., 2010). But, due to timing constraints, trials were shortened to 60 seconds during the novel study. Because statistical analysis was only conducted between arms within the same study, these sources of error had no impact on the results of the study. 

Despite these sources of systematic error and the fact that no significant difference was found between the Sham and Red Test Subjects one hour after exposure to their respective demonstrators, the hypothesis can be accepted. Zebrafish who followed Red Demonstrators consistently continued escaping through the red escape route, and overall, they escaped significantly faster than Sham Test Subjects one week after learning the behavior from the Red Demonstrators. The only time they did not, researchers noted that they were acting lethargic, which was likely due to the fact that they were not in optimal health conditions and had already undergone the positive control.

Should this study be replicated, future researchers should use the larger testing aquarium and one minute trial times for all arms of the experiment to ensure the Test Subjects have adequate space to demonstrate their learned behavior and to maximize timing efficiency. Moreover, researchers should ensure that all Test Subjects are in good health and have sufficient time to recover from the controls before conducting experimental trials, such as a three hour rest period before the first experimental trial in the novel study. 

Social learning is an essential part of zebrafish’ survival mechanisms. It allows members of a shoal to gather information regarding their environment, including food acquisition and predator avoidance, by observing their companions behavior and copying it if beneficial (Zala et al., 2012). As shown by the present novel study, zebrafish are capable of remembering and continuing to replicate these socially learned behaviors over longer periods of time.

Because social learning is a widespread behavior utilized by humans, and particularly children, it is imperative to have a strong understanding of it (Editors of Encyclopedia Britannica, 2019). Social learning has the potential to be used as a tool to better society, as it has been proposed that it can help prevent negative behaviors such as aggression and crime, and enforce positive ones, like effective workplace habits and advocating for social change. However, there were significant knowledge gaps regarding the relationship between social learning and time. Therefore, this novel study offers critical information to the world of behavioral and neuroscience. It confirms the hypothesis that zebrafish are able to continue replicating socially learned behaviors over longer periods of time—showing how impactful social learning is to individuals. One knowledge gap that prevails in this study and should be addressed in future research is the impact of reinforcement over time on social learning. To investigate this further, researchers could add a third and fourth experimental group: Red and Sham Test Subjects who are reexposed to their respective demonstrators daily over the course of the week of experimentation.

SUPPLEMENTAL INFO AND FIGURES

Data Tables 

Lab Notebook

Junior Year Lab Notebook #1 

Junior Year Lab Notebook #2

Junior Year Lab Notebook #3  

Senior Year Lab Notebook #1

Senior YEar Lab Notebook #2  

Senior Year Lab Notebook #3 

WORKS CITED

Adams, M. M. (2018, November 1). Zebrafish—a model organism for studying the neurobiological mechanisms underlying cognitive brain aging and use of potential interventions. National Library of Medicine. Retrieved May 5, 2023, from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6221905/

Avdesh, A. (2012, November 18). Regular care and maintenance of a zebrafish

     (Danio rerio) laboratory: An introduction. National Library of Medicine.

     Retrieved December 17, 2023, from https://www.ncbi.nlm.nih.gov/pmc/articles/

     PMC3916945/

Avdesh, A. (2012, November 18). Regular care and maintenance of a zebrafish

     (Danio rerio) laboratory: An introduction. National Library of Medicine.

     Retrieved December 17, 2023, from https://www.ncbi.nlm.nih.gov/pmc/articles/

     PMC3916945/

Boldness definition and meaning. (n.d.). Dictionary.com. Retrieved June 6, 2023, from https://www.dictionary.com/browse/boldness

Britannica, T. Editors of Encyclopaedia (2019, September 13). Social learning. Encyclopedia Britannica. https://www.britannica.com/science/social-learning

Cambridge Cognition. (2015, August 19). What is cognition? Cambridge Cognition. Retrieved October 18, 2023, from https://cambridgecognition.com/what-is-cognition/

Clark, K. J. (2012, October 12). Stressing zebrafish for behavioral genetics. National Library of Medicine. Retrieved April 15, 2023, from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3470424/

Colwill, R. (2005, August). Visual discrimination learning in zebrafish (Danio rerio). Science Direct. Retrieved May 5, 2023, from https://www.sciencedirect.com/science/article/pii/S0376635705001051

EdX. (2021, June). Theoretical approaches in social work: social learning theory. Social Work License Map. Retrieved May 5, 2023, from https://socialworklicensemap.com/social-work-resources/social-learning-theory-and-its-importance-to-social-work/#:~:text=Applications%20of%20the%20social%20learning,and%20behaviors%20that%20harm%20them

EdX. (2022, February). Introduction to social learning theory in social work. Online Masters of Social Work. Retrieved October 18, 2023, from https://www.onlinemswprograms.com/social-work/theories/social-learning-theory/#:~:text=Applications%20of%20Social%20Learning%20Theory&text=Researchers%20can%20use%20the%20theory,behaviors%20and%20promote%20social%20change.

Filosa, A. (2016). Feeding state modulates behavioral choice and processing of prey stimuli in the zebrafish tectum. Science Direct. Retrieved April 15, 2023, from https://www.sciencedirect.com/science/article/pii/S0896627316300034

Finazzo, V. (2019). Social behavior of zebrafish (Danio rerio) and the expression of the stress hormone cortisol. Long Island University Digital Commons. Retrieved April 15, 2023, from https://digitalcommons.liu.edu/cgi/viewcontent.cgi?article=1056&context=post_honors_theses

Gerlai, R. (2023, January). Zebrafish (Danio rerio): A newcomer with great promise in behavioral neuroscience. Science Direct. Retrieved April 14, 2023, from https://www.sciencedirect.com/science/article/pii/S0149763422004675

Ghoshal, A. (2021, January 13). Group size and aquatic vegetation modulates male preferences for female shoals in wild zebrafish, Danio rerio. Scientific Reports. Retrieved May 5, 2023, from https://www.nature.com/articles/s41598-020-80913-x#:~:text=Introduction,to%20a%20few%20hundred2

Hariharan, S. (2024 (in progress), January). Developmental toxicity and neurobehavioral effects of sodium selenite and selenium nanoparticles on zebrafish embryos. Science Direct. Retrieved December 17, 2023, from     https://www.sciencedirect.com/science/article/pii/S0166445X23003934

How to be a catalyst for social learning [Illustration]. (2014). Association for Talent Development. https://www.td.org/insights/how-to-be-a-catalyst-for-social-learning

Kauleff, A. V. (2013, March). Towards a comprehensive catalog of zebrafish behavior 1.0 and beyond. National Library of Medicine. Retrieved April 14, 2023, from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3629777/

Kit Tan, J. (2022, October 30). Zebrafish: a pharmacological model for learning and memory research. National Library of Medicine. Retrieved May 5, 2023, from https://pubmed.ncbi.nlm.nih.gov/36364200/

Leading Effectively Staff. (2022, April 24). The 70-20-10 rule for leadership development. Center for Creative Leadership. Retrieved June 6, 2023, from https://www.ccl.org/articles/leading-effectively-articles/70-20-10-rule/#:~:text=The%2070%2D20%2D10%20rule%20reveals%20that%20individuals%20tend%20to,10%25%20from%20coursework%20and%20training.

Leff et al. (2009). Chapter 40 - aggression, violence, and delinquency. Science Direct. Retrieved June 6, 2023, from https://www.sciencedirect.com/topics/psychology/social-learning-theory

Lindeyer, C. (2010). Social learning of escape routes in zebrafish and the stability of behavioural traditions. Science Direct. Retrieved May 5, 2023, from https://www.sciencedirect.com/science/article/pii/S000334720900582X

Martin-Soleach et al. (2007, January 8). Appetitive conditioning: neural bases and implications for psychopathology. National Library of Medicine. Retrieved June 6, 2023, from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2693132/#:~:text=Appetitive%20conditioning%20is%20a%20form,their%20association%20with%20a%20reward.

Meshalkina, D. A. (2017, August). Understanding zebrafish cognition. Science Direct. Retrieved April 7, 2023, from https://www.sciencedirect.com/science/article/pii/S0376635716303825

Podila, S., & Zhu, Y. (2017, October). Animation [Illustration]. Wiley Online Library. https://onlinelibrary.wiley.com/doi/abs/10.1002/cav.1866

Rosa, L. V. (2022, September 26). Acetic acid-induced nociception modulates sociability in adult zebrafish: Influence on shoaling behavior in heterogeneous groups and social preference. Science Direct. Retrieved May 19, 2024, from https://www.sciencedirect.com/science/article/abs/pii/S0166432822002972

Roy, T. (2017, November). Social learning in a maze? Contrasting individual performance among wild zebrafish when associated with trained and naïve conspecifics. Science Direct. Retrieved December 17, 2023, from https://www.sciencedirect.com/science/article/pii/S0376635717300451

Santacà, M. (2021, November). Stimulus characteristics, learning bias and visual discrimination in zebrafish (Danio rerio). Science Direct. Retrieved May 5, 2023, from https://www.sciencedirect.com/science/article/pii/S0376635721001832

Shimada, Y. (2022, October). Zebrafish: A pharmacological model for learning and memory research. National Library of Medicine. Retrieved December 17, 2023, from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9657833/

Sinauer Associates. (2013). Critical Values of the Chi-Square Distribution with d Degrees of Freedom. Sinauer Associates. Retrieved May 16, 2024, from https://www.mun.ca/biology/scarr/4250_Chi-square_critical_values.html

Slabbekoorn, H. (2020). Zebrafish personality, stress physiology and behaviour in the context of sound exposure. Universiteit Leiden. Retrieved April 15, 2023, from https://www.universiteitleiden.nl/en/research/research-projects/science/ibl-zebrafish-personality-stress-physiology-and-behaviour-in-the-context-of-sound-exposure#:~:text=Active%20swimming%2C%20throughout%20the%20tank,tank%20walls%20and%20close%20together.

Sterelny, K. (2019, December 13). Humans as model organisms. The Royal Society Publishing. Retrieved May 5, 2023, from https://royalsocietypublishing.org/doi/10.1098/rspb.2017.2115

Triplett, R. (2015, October 26). Crime, social learning theory of. Wiley Online Library. Retrieved June 6, 2023, from https://onlinelibrary.wiley.com/doi/abs/10.1002/9781405165518.wbeosc157.pub2#:~:text=The%20social%20learning%20theory%20of,commit%20crime%20nor%20to%20conform.

UC Berkeley. (2023). What is social learning (and how to adopt it). UC Berkeley. Retrieved April 15, 2023, from https://hr.berkeley.edu/how-social-learning-theory-works

UMass Dartmouth. (n.d.). Nervous System Final. Retrieved May 5, 2023, from https://www.umassd.edu/media/umassdartmouth/center-for-indic-studies/acad_fall08_coursematerial.pdf

Vergauwen, L. (2010, October). Long-term warm or cold acclimation elicits a specific transcriptional response and affects energy metabolism in zebrafish. Science Direct. Retrieved December 17, 2023, from https://www.sciencedirect.com/science/article/pii/S1095643310003454

Washington University. (2022, August 2). Policy for the housing and care of zebrafish and other fish species in satellite facilities. Washington University. Retrieved December 17, 2023, from https://research.wustl.edu/app/uploads/2017/05/Housing_and_Care_of_Zebrafish_in_Satellite_Facilities.pdf

Western Governers University. (2020, May 15). How social learning theory works in education. Western Governers University. Retrieved May 5, 2023, from https://www.wgu.edu/blog/guide-social-learning-theory-education2005.html#close

Wong, K. (2010, April). Analyzing habituation responses to novelty in zebrafish (Danio rerio). Science Direct. Retrieved April 15, 2023, from https://www.sciencedirect.com/science/article/pii/S0166432809007499

Zala, S. M. (2012, June). Different social-learning strategies in wild and domesticated zebrafish, Danio rerio. Science Direct. Retrieved April 8, 2023, from https://www.sciencedirect.com/science/article/pii/S0003347212001571

Zebrafish International Resource Center. (n.d.). ZIRC frequently asked questions. Zebrafish International Resource Center. Retrieved December 17, 2023, from https://zebrafish.org/documents/faq.php#9