This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
In recent years, Trichomonas gallinae was reported as a rapidly growing cause of mortality among garden and wild birds in Britain. We attempted to isolate this parasite from the wild birds with intention to study its viability outside the host body in various environmental conditions imitating the garden bird baths (tap or rain water at different temperatures). Due to the difficulties with isolation of parasite from the wild bird material available, we have performed the viability studies on the closely related microorganism, Trichomonas vaginalis.
KEY WORDS: Avian trichomonosis, Trichomonas gallinae, Trichomonas vaginalis, birds mortality, bird epidemics.
In contrast with previous reports, T.gallinae was very hard to isolate from the birds, especially during the winter period. It is likely that parasite is more common during the breeding season when it can spread easily by mouth to mouth feeding.
Closely related microorganism T. vaginalis survives better at higher temperatures and in tap water rather than rain water.
Due to the close similarity between two species of Trichomonas, it can be assumed that T. gallinae would show similar viability trends. Significantly better host-free survival at higher temperatures can help to explain the seasonal character of T. gallinae infections.
Since 2005, avian infections caused by T. gallinae are on the increase in the UK. T. gallinae is a parasite of the upper digestive tract of pigeons and doves that causes avian trichomonosis. The usual hosts are doves and pigeons but in the UK the parasite has recently spread to small songbirds such as finches and passerines and now poses a serious threat to the number of wildlife birds. Robinson and colleagues (Robinson et al., 2010) reported trichomonosis as the main factor of British finches mortality in 2005 and 2006. This is the first time the trichomonosis epidemic was reported for the non-columbiform species of birds. It confirms the ability of this protozoan parasite to move successfully from one species of host to another and to jump the avian taxonomic groups, with dramatic consequences for the abundance of affected species.. The British Trust for Ornithology estimates that 500,000 finches died from this parasitic disease in 2007 alone (British Trust of Ornithology, 2007). This increase in mortality is believed to be connected to the increase in population of wood pigeons, major carriers of infection. Also, due to the extensive usage of land and habitat degradation the gardens are commonly used as refuge by wild birds (Toms, 2007). The feeders and water baths in the gardens are often shared by wild species and pigeons. 48% of British gardens provide food for the wildlife artificially (Davies, Fuller, Loram, Irvine and Sims et al. 2009).
Although T. gallinae is more common in pigeons and doves, many cases of trichomonosis in other species of birds have been reported worldwide (Forrester and Foster, 2008; Anderson, Grahn, Van Hoosear and Bondurant, 2009; National Wildlife Health Center, 2002). Previously, trichomonosis has pushed the passenger pigeon Ectopistes migratorius to extinction (Stabler 1954), and endangered the survival of several other species such as island-endemic pink pigeon Nesoenas mayeri (Bunbury, Jones,Greenwood and Bell, 2008) and Bonelli's eagle Hieraaetus fasciatus (Höfle, Blanco, Palma and Melo, 2000; Real, Mañosa and Muñoz 2000). This makes trichomonosis a main concern for conservation. The Royal Society for the Protection of Birds (RSPB) has listed trichomonosis as one of the five most common diseases in garden birds.
Outbreaks of trichomonosis usually occur every year during late summer and autumn. In the parts of the country suffering the most serious outbreaks, the population of greenfinch has been recorded to drop by a third, and chaffinch populations by a fifth. The parasite is thought to be transmitted between species via infected drinking water shared by pigeons and songbirds, typically garden birdbaths and feeders.
To advise any approach aimed to prevent the spread of the disease among the garden birds, we have to have clear understanding of the biology of causative microorganism, its life cycle and transmission mechanism. There are some gaps in the knowledge here, which this work aims to address.
The biology of Trichonomas gallinae
T. gallinae is a flagellate protozoan from the order Trichonomidae. This is a parasite of the upper digestive and respiratory tracts of birds, most commonly pigeons and doves, although a variety of other species such as chickens, turkeys, birds of prey, parrots, and canaries can be affected. This protozoan lives and feed on mucosal surfaces. Most infections caused by T. gallinae are asymptomatic, with the individuals acting as carriers. Birds that do show symptoms have necrotic ingluvitis lesions in their throats, drool saliva, are lethargic with fluffed up plumage and swollen throats (Forrester and Foster, 2008; Robertson et al, 2010). Lesions cause their throats to be swollen. Birds find swallowing of food and water intensely difficult and eventually starve or choke to death within 4-18 days (Cole and Friend, 1999). Lesions can happen in their livers or lungs.
Like most Trichomonidae species, T. gallinae feeds on the cell debris and bacteria on the mucosal surfaces. More virulent strains, however, start to break down and consume the normal host tissues and cells, causing injuries and inflammatory response. This leads to pain, itching and swelling. In severe cases, the parasites eat through the lining of digestive or respiratory tracts causing injuries to underlying tissues.
The parasite can be transferred through direct contact between birds during courtship and regurgitant feeding (transfer to young birds from their parents). Indirect routes are possible such as through drinking water and food contaminated by regurgitated saliva and food particles or via contaminated bill and mouth (Cole and Friend 1999; Forrester and Foster, 2008; Bunbury, Jones, Greenwood and Bell, 2007).
Unlike many other single cell parasites, T. gallinae does not have a resistant cyst stage in its life cycle, and therefore can be transferred predominantly via close host-to-host contact. The absence of cysts makes T. gallinae vulnerable to desiccation in harsh environmental conditions. Parasite can only survive for short periods outside the avian host. Nevertheless, since water route is a key to its transmission, the parasite must survive for some time in media such as contaminated water or food. Water can be easily contaminated when infected birds defecate into the communally used water sources such as garden bird baths. The extent of parasite's ability to survive outside the host's body is not well studied.
Survival of Trichonomas species in aqueous media
Very few studies were conducted with the aim of studying the survival of Trichomonas gallinae in aqueous media. The closely related human parasite, Trichomonas vaginalis, was studied much better in this respect. Trichomas vaginalis is also a species of flagellated microorganism which comes from the same genus Trichomonas. T. vaginalis is usually found on the mucosal surfaces lining the female reproductive tracts in humans and cattle. All organisms in this group are anaerobic, flagellated protists that usually have 3-5 free flagella (long hair-like structures protruding from the body) that help them to swim. A thicker posterior flagellum passes backwards along the side of the body forming the undulating membrane called axostyle. Parasites are usually pear-shaped and are 10-12 microns in size. A single nucleus is located at the round end of the parasite. The parasites reproduce by longitudinal fission dividing into two along its axis.
Trichomonas vaginalis is common in humans and can be transmitted sexually. According to Gerbase et al. (1998), up to 170 million people become infected annually with this parasite. Women are the most common victims (Young 2006), whereas men usually do not show signs of the disease (Kreiger, 1995). Although Trichomonas is a sexually transmitting species, it can be transmitted non-sexually as well by sharing of douche nozzles, specula, toilet seats, and also through the swimming pool water (Catterall and Nicol, 1960; Piekarshi and Saathoff, 1973; Whittington, 1957). Catterall and Nichol (1960) reported that T. vaginalis could survive in the swimming pool for several hours. Piekarski and Saathoff (1973) further confirmed the reports of infections of T. vaginalis in the swimming pool. Some researcher (for instance, Nett and Schar (1986)) argued that T. vaginalis is unlikely to be infectious after being exposed to the pool. However, a recent study published in 2008 reported that T. vaginalis can remain viable and infective in the swimming pool water for few hours, depending on its strain infectivity. This is also possible if the T. vaginalis is cultured long-term or freshly isolated (Pereira-Neves and Benchimol 2008).
Aim of this study
In this project we want to determine how long the parasite can survive in water representative of bird bath water. We will incubate the culture-grown organisms in water under different conditions (different pH and temperature) and monitor the survival over time by the counting of viable parasites. We will isolate a parasite culture from the upper digestive tract of wood pigeons that have been shot for human consumption and obtained from local butchers, farmers and game hunters. We will examine the parasite's survival over time in water representative of bird bath water (tap and rain water), using a range of temperatures. The parasite's survival rate will be measured by counting viable parasites under the phase contrast microscope. To establish the experimental techniques, we will perform experiments on closely related organism, Trichomonas vaginalis, which we can obtain from another laboratory.
We expect that parasite is capable to stay alive and infectious in aqueous media for at least some time. It is also likely that the survival rate depends on the exact conditions (pH and temperature). Temperature dependence of T. gallinae survival in water might explain the observed seasonal patterns of trichomonosis infection among birds. The results of this study might help to conclude whether and to which degree the garden baths water can indeed mediate the spread of disease among the birds.
MATERIALS AND METHODS
I) Culture media and reagents
MDM (modified diamonds medium) was prepared according to previously described procedure (Cover et al., 1994). Horse serum was obtained from Langford, Bristol. Antibiotics (penicillin and streptomycin) were supplied by Sigma.
Incubation media contained 90 ml of MDM, 10 ml of heated horse serum, and 1ml of antibiotics mixture (penicillin and streptomycin).
II) Tap water and rain water
Both types of water are commonly used in bird bath. The tap water was collected from the laboratory tap, and the rain water was obtained outside the 3rd year lab. Both water samples were left for a day to be de-chlorinated. The pH of water was measured using indicator strips for pH range from 5 to 10. The strips need to be immersed in water and kept there until there is no further color change (approximately 2 min). The pH of tap water was 8.5, and rain water had pH 5.
Phase contrast microscope Laborlux 12 (Leitz) (Wetzlar Germany 153558) was used to examine the samples and make quantitative estimations of parasite numbers.
Mini-centrifuge Mini Spin (Eppendorf UK Ltd., UK) was used to centrifuge the parasite most effectively from mixed medium.
Clicker (Trumeter co. ltd, Manchester, UK) and haemocytometer (Weber Scientific International, Hawksley Technology, Sussex, UK) were used to count parasites during microscopic examination of samples.
Compenstat-control thermostats (Gallenkamp, UK) set at certain temperatures were used to incubate the parasite containing samples.
I) Isolation of Trichomonas gallinae
For the isolation of Trichomonas gallinae, the head and neck of 24 wood pigeons (supplied after preparation of the bird for human consumption) were obtained from Ruby and White Butchers (Whiteladies road, Bristol, Avon), as well as from several woods near Tring, Hertfordshire, UK. Only the birds killed within 24 hours before sampling were used to ensure reliability of data (Erwin et al 2000).
Each sample was first visually examined for presence of the yellow cankers or lesions or any other signs of infections. None of the birds showed visual signs of infection. The birds without any visual signs were also used since most infected birds are asymptomatic (Forrester and Foster 2008; Robertson et al 2010).
The mouth and gullet of the pigeons were wiped with a sterile swab and inoculated into liquid culture medium containing antibiotics (Trichomonas medium no.2 and CPLM Trichomonas broth; both in containers of 5 ml) and incubated at 37oC in anaerobic conditions for 48 h. Both of the medium are reported to grow Trichomonas species well since they contain antibacterial compounds to prevent contamination (Huso et al 2011; Chaudhari and Singh 2011).
Cultures were monitored for growth of parasites and sub-passaged into fresh medium as required to establish a cultured line of T. gallinae.
After 48 hours, 10µl of the culture is placed onto a microscope slide with a cover slip and viewed under 10x and 40x magnification to search for alive Trichomonas gallinae.
II) Inoculation of Trichomonas vaginalis
The metronidazole-susceptible G3 strain of T. vaginalis was obtained from Natto and coworkers (Natto et al., 2012) in frozen state. The leftovers after inoculations were kept frozen in liquid nitrogen to keep microorganisms virulent (Diamond et al., 1965). To inoculate the T. vaginalis, it has to be defrosted slowly, to prevent heat shock and killing of the parasites. The samples (200 µL) were inoculated into bijou tubes with culture medium (5 ml) which were incubated at 37oC. Inoculation is done with 1:25 dilution ratio in the tube. This is the most efficient ratio to avoid overpopulation and subsequent killing of the parasites. There were in average 100 parasites in the counting cell. It is necessary to sub-passage the strain into fresh medium at least once every 3 days. T. vaginalis was used as a contingency plan unless T. gallinae is found through inoculation and incubation.
III) Measurement of T. vaginalis survival in tap and rain water.
200 µL of parasite-containing inoculated medium were placed into Eppendorf tubes (one tube for each incubation temperature). Tubes were centrifuged for 3 minutes at 5,000 g. Centrifugation at this speed and duration allows to precipitate the parasites to the bottom of the tube without killing them. After centrifugation, 40 µL of inoculated medium were collected from the bottom of centrifuged tube and put into fresh tubes with 460 µL of either tap or rain water (one tube for each temperature point). The tubes were incubated at four different temperatures: 4, 10, 20 and 37oC. Survival rates were measured against time. After appropriate incubation period (0, 30, 60, 90, 120 or 150 min), the tubes were centrifuged, and 10 µL samples were taken from the bottom of the tube for microscopic investigation. Samples were pipetted on to the edge of coverslip which is covered on the hemocytometer to count the number of T. vaginalis alive using a phase contrast microscope. The numbers were counted from 16 squares in any corner of the huge square using the clickers. The parasites were considered to be alive when the cell wall was intact and mobile tail was present. Dead parasites had dispersed cell wall. Using the numbers of alive and dead parasites, the survival rate percentages were calculated. These numbers reflect the T. vaginalis viability.
IV) Further testing of viability of T. vaginalis survived in water
In order to check if the trichomonads that survived in water were still cytotoxic, co-incubation of the parasites with primary cell cultures was performed and their cytotoxicity was compared with parasites routinely grown in culture media. Modified protocol of Pereira-Nunes et al. (2008) was used for these experiments.
At each time point of incubation of T. vaginalis in water, 0.5 ml samples of water contaminated with parasite were centrifuged at 5000 g for 3 min. 40 µL pellet from the tube bottom was transferred to fresh MDM culture medium (1 mL) and maintained at 37 °C for 24 h in order to determine whether these cells were able to proliferate. The presence of T. vaginalis in the culture medium after this incubation period indicated that parasite was able to proliferate and therefore remained viable and infective.
V) Statistical analysis
Two sets of data collected from the same experiment were transformed into arc-sin and further tested with Q-Q plot (the transformed data was again transformed into residuals to make the Q-Q plot). Graphically, all the points were close to the line y = x, concluding strong normality for all data (see the graphs in Appendix A).
The data were subjected to paired Student's t-tests, two-way analysis of variance (ANOVA) and Tukey Post-Hoc Test.
Two-way analysis of variance was used to see the effect of type of water and temperature (two fixed factors) on survival rate (dependent variable).
If there was a significance in the effect of type of water on survival rate, paired t-test was further used to compare the effect of water type on survival rate of parasites.
If there was a significance in the effect of temperature on survival rate, Tukey Post-Hoc test was further used to examine which specific temperatures are significantly different from each other in both types of water.
If both temperature and water type have effect on survival rate, interaction plot is produced to visualize the relationship between the temperature and the type of water and see whether the interaction between temperature and water type is significant.
Isolation of T. gallinae
Unfortunately, our attempts to isolate T. gallinae from the wild birds were unsuccessful. None of the birds had visible signs of infection, but this is often the case when the disease is still present. The methodology used for isolation of T. gallinae was tested by previous researchers, and therefore it is unlikely that the lack of success is connected with the use of unreliable methods.
Due to the failure in obtaining T. gallinae, we decided to investigate the survival of closely related organism, T. vaginalis, using the culture of parasite obtained from another laboratory. From December 2012 to January 2013, we have done a total of 40 experiments, 5 repeats for each temperature and type of water.
Effect of temperature on survival of T. vaginalis
The results of experiments performed in two different types of water clearly demonstrate that survival of T. vaginalis improves with the increase of temperature (Fig. 1 and 2). In the temperature range investigated, the optimum temperature for T. vaginalis survival was 37oC. At this temperature, up to 20% of parasites were still alive even after 150 hours of incubation. Unlike other temperatures, parasites in 4oC drastically dropped from the start.
The temperature has stronger significant effect on the survival rate of T. vaginalis (two-way ANOVA test, F=41.345, df=3, p>0.001; Appendix B) than the water type does (two-way ANOVA test, F= 17.266, df=1, p>0.001; Appendix B). However, there was no effect of temperature and water type on the T. vaginalis survival (two-way ANOVA test, F=0.669, df=3, p=0.571; Appendix B). Survival rate of T.vaginalis decreases dramatically from 4 oC to 20 oC (Tukey Post-Hoc Test, df=15, p< 0.05; Appendix B) but not from 20 oC to 37 oC degrees (Tukey Post-Hoc Test, df=15, p>0.05; Appendix B). The survival rate reaches its minimum when the T. vaginalis is put in the water with 20 oC to 37 oC.
Fig. 1. Influence of temperature on survival of T. vaginalis in tap water. Error bars represent 95% confidence intervals.
Fig. 2. Influence of temperature on survival of T. vaginalis in rain water. Error bars represent 95% confidence intervals.
Effect of water type on survival of T. vaginalis
Figs. 3-6 show the survival of T. vaginalis in two different types of water at four different temperatures. The red lines represent the survival in tap water, while the blue lines show survival in rain water. The graphs show that in every case T. vaginalis was able to survive longer in tap than in rain water. As temperature increases, the difference between tap and rain water increases. Two-way ANOVA test ((F= 17.266, df=1, p>0.001) and paired t-tests (figure 3-6, p<0.05) confirmed the survival rate of T.vaginalis is higher in tap water than the rain water.
Fig. 3. Effect of water type on survival of T. vaginalis at 4oC. Error bars represent 95% confidence intervals (t=3.406; df=15; p=0.004). Survival rate in tap water is higher than in rain water.
Fig. 4. Effect of water type on survival of T. vaginalis at 10oC. Error bars represent 95% confidence intervals (t=3.647; df=15; p=0.002). Survival rate in tap water is higher than in rain water.
Fig. 5. Effect of water type on survival of T. vaginalis at 20oC. Error bars represent 95% confidence intervals (t=7.337; df=15; p>0.001). Survival rate in tap water is distinctly higher than in rain water.
Fig. 6. Effect of water type on survival of T. vaginalis at 37oC. Error bars represent 95% confidence intervals (t=6.708; df=15; p>0.001). Survival rate in tap water is distinctly higher than in rain water.
We have measured the maximum survival time for parasite incubated in aqueous media at different conditions (Table 1). Although only one repeat was made (due to time constraints) and the survival time at 10 degrees seems to be a bit high and out of line with other data, there is overall a trend towards longer viability at warmer temperatures. The results of the viability test suggest that if cells are able to remain viable then they are likely to still be infective.
Table 1. Viability results. Maximum survival time of parasites in different conditions.
Tap water, time until death (min)
Rain water, time until death (min)
Isolation of T. gallinae
The original intention of this research project was to isolate T. gallinae from wild birds and study the survival of this parasite in various environmental conditions. Unfortunately, our attempts to isolate T. gallinae were unsuccessful which prompted us to do the work on the closely related microorganism, T. vaginalis.
One of potential reasons for the failure to isolate T. gallinae might be the high mortality of parasite upon storage. Also, there is a possibility that the birds were kept in the fridge before the samples were collected from them. Previous research (Erwin et al., 2000) shown that T. gallinae survives in reliable numbers for 8 hours in the carcasses of white-winged doves. Parasite's survival gradually declines in the carcasses upon storage, but even after 48 h it can be detected in 44% of positive birds. Survival and viability of T. gallinae depends significantly on the storage conditions, with higher temperature contributing to lower detection rate (Erwin et al., 2000). To increase our chances of finding the parasite, we repeated the isolation procedure using birds that were shot the same afternoon. Unfortunately, no birds showed any infection symptoms and no T. gallinae was isolated in these experiments either. We can only assume that due to the seasonal nature of infection the chances of getting an infected bird were simply too low during winter, or the infection is not very common in our geographic region. The parasite is most likely to be more common during the warm breeding season when it can spread easily by mouth to mouth feeding.
Despite clear difference between T. gallinae and T. vaginalis in terms of hosts and infections caused, both organisms are closely related, and it is highly likely that any trends observed for T. vaginalis will be similar to those in the pathogen affecting the garden birds.
High death of birds due to salmonellosis, not T.gallinae?
The fact that we failed to find any signs of trichomonosis infection in our samples makes us speculate that this infection is not as common as the resent statistics makes us think.
Opportunistic monitoring of garden bird mortality by the Royal Society for the Protection of Birds (RSPB) between 2001 and 2004 showed an annual seasonal mortality peak in mid-winter (December - January), with 37-76% of reports per annum occurring in these two months. Post mortem examinations indicated that this seasonal peak was largely due to salmonellosis in Fringillidae and Passeridae species. No trichomonosis cases were observed during these studies.
During the next several years, however, the number of reported trichomonosis cases grew dramatically. In 2006 they already comprised 50% of all reported incidents of garden bird mortality. This statistics, however, is based primarily on the unsolicited reports by the public. Sick and dead birds at affected locations were usually observed in close vicinity to garden bird feeding stations and were showing some unspecific signs of disease, such as lethargy and fluffed-up plumage, often in combination with dysphagia. A number of other diseases and conditions different from T. gallinae infections (candida, bacterial infections, avian pox viruses etc.) can produce similar manifestations such as white or yellow plaques and ulcers in the mouth. Therefore, the reported numbers for T. gallinae infection related mortality should be treated with caution. It would be also reasonable to suggest that the increased mortality from trichomonosis reported in the previous years is not necessarily an annual occurrence, and also there should be a significant geographical variability in the spread of disease. The report of the 2006 epidemics (Robinson et al., 2010) highlighted a significant geographical variability in the disease distribution on the British Isles. Also, the real extent of the T. gallinae epidemic is hard to estimate. The calculations of the percentage of infected birds can be affected by sampling size, season, detection method used and many other factors (Robinson et al., 2010). The data for birds abundance in the particular area should also be taken into account when estimating the epidemics extent. Besides, not all techniques successfully identify the presence of infection in birds. Molecular detection of infection based on the nested PCR protocol provides more reliable identification. This method, however, was not employed for the purpose of this study.
We can hypothesize that significant number of bird deaths might currently occur due to salmonellosis rather than trichomonosis. Without proper laboratory-based study, however, this hypothesis will remain highly speculative.
Survival of T. vaginalis in different environmental conditions
All our data including the viability data for T. vaginalis clearly show a trend for better survival at warmer temperatures. They also demonstrate that this parasite survives better in tap water. These conclusions are supported by the statistically significant data. Due to the time constraints in our experimental work, the conclusions from viability test were based on the results of only one set of experiments, but the results overall gave a clear picture of trend towards longer viability at warmer temperatures except the survival time at 10 degrees which might reflect some unidentified experimental mistakes. Nonetheless, the analysis of pseudo-replicates (different samples from one culture instead of collecting T. vaginalis from different individual hosts) lacks certain degree of statistical significance and would need to be confirmed by further studies.
The results of the viability test suggest that if cells are able to remain present in the medium then they are likely to still be infective, although we would need live tissue samples to be certain of this. This opinion is indirectly confirmed by the observations of Pereira-Neves and Benchimol (2008) who found that T. vaginalis survived in swimming pools are able to adhere and disrupt bovine oviduct epithelial cell (BOECs) monolayers.
As temperature and water type significantly influenced the survival of T. vaginalis, this can imply that T. gallinae would also survive longer at higher temperatures, as well as in tap rather than rain water.
We can therefore predict that T. gallinae would survive longer in the summer months, which would result in higher infection and prevalence at this time. Our data seem to be in agreement with the observed seasonal pattern for T. gallinae infections in the UK.
We also examined a sample from another rain-filled water bath in which we observed a number of other organisms present such as Haematococcus pluvialis, Ceratium sp., Chrysococcus sp., as well as hydra and various dinoflagellates. The competition for the same niche, which is the rain-filled water, must have driven T.gallinae to not survive well.
Using Universal Indicator paper, we determined the pH of water used in these experiments. Tap water had pH 8.5, and rain water pH 5.0. Previous studies have shown that the optimum pH for T. vaginalis proliferation is between 5.45 and 5.55, although it is able to survive in the pH range between 4.9 and 7.55 (Johnson, 1940). This suggests that, as the rain water was closer to the optimum pH for T. vaginalis growth, the pH of water was not the reason for longer survival of T. vaginalis in tap water. Of course, tap and rain water pH varies depending on location, and any effects of pH on our samples may not be the same in another location.
It is not clear at this stage which particular characteristics of tap water were behind the better survival rate of T. vaginalis. One of the possibilities would include the presence of a variety of inorganic ions in the tap water, while rain water has very little amount of inorganic salts.
Obviously, despite the close similarity between two parasites, there is no guarantee that the tendencies observed for survival of T. vaginalis in various environmental conditions will be reproduced for T gallinae. Proper studies with this parasite need to be performed to get the clear insight in the way the avian trichomonosis spreads among the wild birds.
We thanks Dr. Natto and colleagues for providing the isolated samples of metronidazole-susceptible G3 strain of T. vaginalis. I would like to thank Prof. Wendy Gibson for her supervision and support throughout the duration of project. We also would like to thank Gary Barker for his advice on statistics in our Methods section.