Mating motivation: the effect of eliminating the male function on sexual behaviour in a simultaneous hermaphrodite

Difference in male and female motivation to mate can be attributed to con ﬂ icts over investment in gametes and offspring. While such motivations are relatively straightforward in separate-sexed species, they may become intertwined in simultaneous hermaphrodites. Such organisms produce both types of gametes and each sperm donor is also a sperm receiver. For the pond snail, Lymnaea stagnalis , male motivation has been shown to become temporarily low after having donated sperm. Moreover, previous studies indicated that being inseminated can be costly for the sperm recipient's reproductive success because accessory gland proteins transferred in the ejaculate can temporarily lower female and male reproductive success. In addition, potential sperm recipients have been found to exhibit behaviours that suggest that they try to avoid being inseminated. Such behaviours could be motivated via the sperm- receiving, female role in this pond snail. To test this, we experimentally eliminated this species' male role using a simple surgical procedure, thus creating ‘ feminized ’ snails. We subsequently observed mating behaviour and quanti ﬁ ed egg laying and hatching success as a proxy of female reproductive success. Although we did not ﬁ nd an effect of feminization on hatching success, we did ﬁ nd a signi ﬁ cant decrease in the occurrence of ‘ biting ’ behaviour in feminized individuals compared to sham-operated and control snails. Although biting has been proposed to be a female behaviour, it is likely dictated via the male role because feminized hermaphrodites showed less biting, possibly because they no longer had to defend their paternity success. The other prominent recipient behaviour, ‘ crawl-out ’ , was not eliminated when performance of the male role was surgically removed, which suggests that this behaviour

Difference in male and female motivation to mate can be attributed to conflicts over investment in gametes and offspring. While such motivations are relatively straightforward in separate-sexed species, they may become intertwined in simultaneous hermaphrodites. Such organisms produce both types of gametes and each sperm donor is also a sperm receiver. For the pond snail, Lymnaea stagnalis, male motivation has been shown to become temporarily low after having donated sperm. Moreover, previous studies indicated that being inseminated can be costly for the sperm recipient's reproductive success because accessory gland proteins transferred in the ejaculate can temporarily lower female and male reproductive success. In addition, potential sperm recipients have been found to exhibit behaviours that suggest that they try to avoid being inseminated. Such behaviours could be motivated via the spermreceiving, female role in this pond snail. To test this, we experimentally eliminated this species' male role using a simple surgical procedure, thus creating 'feminized' snails. We subsequently observed mating behaviour and quantified egg laying and hatching success as a proxy of female reproductive success. Although we did not find an effect of feminization on hatching success, we did find a significant decrease in the occurrence of 'biting' behaviour in feminized individuals compared to sham-operated and control snails. Although biting has been proposed to be a female behaviour, it is likely dictated via the male role because feminized hermaphrodites showed less biting, possibly because they no longer had to defend their paternity success. The other prominent recipient behaviour, 'crawl-out', was not eliminated when performance of the male role was surgically removed, which suggests that this behaviour is female role motivated. The species and experimental approach used enabled us to disentangle sexual motivations of a simultaneous hermaphrodite and show that not all behavioural components have the same motivational basis.
© 2023 The Author(s). Published by Elsevier Ltd on behalf of The Association for the Study of Animal Behaviour. This is an open access article under the CC BY license (http://creativecommons.org/licenses/ by/4.0/). For many decades biologists have studied the relationship between sexual traits and reproductive success to understand sexual selection. It is now clear that sexual reproduction generally involves a conflict of interest between mating partners (e.g. Kokko & Jennions, 2014). For example, traits that are favoured by sexual selection can enhance an individual's ability to acquire mates, but likely come at a cost for the (reproductive) fitness of its mating partner (see also . This phenomenon of sexual conflict and its effects on selection for sexually dimorphic traits have gained traction, especially since the 1970s, and numerous examples of coevolutionary arms races between the sexes have been discovered since then (Arnqvist & Rowe, 2013). A well-known example of such counter-adaptive coevolution is found in the genital coevolution of waterfowl. The corkscrew penis of male waterfowl enables them to coerce unwanted mating, but females have evolved convoluted vaginal ducts that help avoid full intromission of the penis, thus hampering successful insemination (e.g. Brennan & Prum, 2015).
Sexual conflict theory is currently relatively well understood but, like the majority of sexual selection studies, is largely based on the separate-sexed situation (Arnqvist & Rowe, 2013). Other forms of sexual reproduction, such as simultaneous hermaphroditism, have, for no reason, remained understudied. Simultaneous hermaphroditism is a widespread reproductive strategy in plants and animals (Eppley & Jesson, 2008;Jarne & Auld, 2006;Sch€ arer et al., 2014). Because such individuals produce both types of gametes at the same time, it may seem that sexual selection and conflict are unlikely to occur. However, their gametes are also unequal in size and provisioning capacity, just like in separate-sexed species (e.g. Anthes et al., 2010;Bateman, 1948;Parker et al., 1972). In other words, simultaneous hermaphrodites are also anisogamic and Bateman's principle and sexual conflict seem to apply equally, that is, individuals mainly mate to donate sperm and attempt to avoid receiving sperm (Charnov, 1979;Anthes et al., 2010;Hoffer et al., 2017;but see Fromonteil et al., 2023). Indeed, morphological adaptations that optimize male reproductive fitness, and female counter-adaptations in line with sexual conflict, have already been documented in simultaneously hermaphroditic species (e.g. Koene & Schulenburg, 2005).
While we understand why traits coevolve to minimize sexual conflict, what is lacking from the current sexual conflict framework is a proper integration of how sexual motivation is involved and how this affects reproductive success. We define such motivation as the drive and/or willingness to perform sexual behaviour. This is particularly relevant for hermaphrodites with the simultaneous presence of both sexual functions, because the motivation to perform the male or female role may even be intertwined and/or cause within-individual conflict over which role to prioritize. Obviously, the performance of each sexual role is expected to lead to the optimal use of both functions for maximal total reproductive fitness of the individual. Not all hermaphrodites are suitable for addressing such questions. For example, in hermaphroditic species with reciprocal copulation, it is impossible to disentangle male from female motivation because their mating behaviour forces them to perform both (e.g. Anthes et al., 2010). However, unilaterally mating simultaneously hermaphroditic species offer the unique opportunity to separate male and female motivation experimentally (e.g. Daupagne & Koene, 2020).
One such species is the pond snail, Lymnaea stagnalis, which serves as a model species for various biological disciplines (reviewed in Fodor et al., 2020). This freshwater snail species is simultaneously hermaphroditic and mates unilaterally, which means that during one copulation the mating partners perform one sexual role (i.e. one individual acts as sperm donor and the other as sperm recipient). After insemination the mating partners can swap sexual roles (Koene & Ter Maat, 2005). Which mating role the individual prefers to perform (first), as explained below, is related to the volume of the seminal fluid-producing prostate gland (De Boer et al., 1997). After mating in the male role, and thus donating sperm and seminal fluid, the prostate gland is partially depleted and it takes up to 8 days of abstinence for the gland to be fully replenished . During copulation the donor transfers sperm and seminal fluid (i.e. accessory gland proteins from the prostate gland) to the recipient, which can store and use sperm from a single mating for up to 2 months (Nakadera, Blom et al., 2014).
Using this species, Hoffer et al. (2010) performed an experiment in which snails were 'feminized' (i.e. surgically made female-acting only) by removing their motivation to perform the male role. This was based on an earlier finding by De Boer et al. (1997) showing that in L. stagnalis a nervous connection exists between the prostate gland and the central nervous system (CNS). This informs the animal about how full the prostate gland is; if the gland is refilled after donating seminal fluid (in a previous mating) this connection provides a permissive signal to the CNS that results in the animal becoming motivated to mate in the male role again. Surgically severing this nervous connection permanently removes the animal's information about whether to mate as male and as a result it no longer mates as a male and only performs the female role (De Boer et al., 1997;De Visser et al., 1994;. This surgical procedure allowed researchers to partially disentangle the costs and benefits of mating in the male and/or female role and Hoffer et al. (2010) showed that pairs of snails that were both feminized had a higher fecundity (measured in terms of egg production, in this case fertilized with stored sperm from previous mating partners) than individuals that were restricted to performing one sexual role. This suggests that freed-up reproductive resources from not having to perform the male role were reallocated to the remaining, female role. If only one individual in the pair was restricted to the female role, however, such individuals laid similar numbers of eggs as intact hermaphrodites performing both roles (note that the same applied in reverse, when an individual could only mate in the male role because it had a feminized partner).
To get a better understanding of how reproductive resources are invested and affect the individual's fitness and its mating partner(s), it is essential to take sexual behaviour into account. For example, Moussaoui et al. (2018) reported that recent insemination of L. stagnalis correlates with reluctance of the recipient in subsequent mating interactions to mate in the female role, resulting in the recipient expressing behaviours that seem to be aimed at discouraging the new mating partner (e.g. crawling above the water line and biting the potential mating partner). Such discouragement of a new partner mounting the recipient's shell was referred to as 'avoidance behaviour'. Recent follow-up experiments suggest that this avoidance behaviour may be induced by the receipt of accessory gland proteins and/or spermatozoa and may be used by sperm recipients to discourage superfluous inseminations (Daupagne & Koene, 2020).
Behavioural indications of avoidance of reinsemination, which may be costly due to the receipt of accessory gland proteins , hint at a shift in motivation to mate in the female role. Therefore, this study aimed to examine whether these 'avoidance behaviours' are indeed female role motivated. We tested this hypothesis by surgically creating individuals that lack the nervous connection informing them about their male motivation status. For such individuals that are restricted to the female role (feminized), we expected a shift in sexual motivation towards the female role and thereby a reduction in the previously reported 'avoidance behaviour'. If some of these behaviours do not change compared to the control groups, this implies that male role motivation also plays a role in their expression. In addition to the behavioural observations, we quantified egg production as well as hatching success to test whether feminization resulted in higher female reproductive fitness in this simultaneous hermaphrodite. Based on Hoffer et al. (2010), we expected that egg numbers would not change, while hatching success has not been measured before in this context but has been shown to change in longer term experiments with this species (Hoffer et al., 2017).

Housing
Adult snails, 3 months old, measuring on average 2.92 cm (± 0.022 SEM) in shell length, were collected from our agesynchronized, mass-rearing culture. Throughout the experiment, except during the mating observations, we kept the snails individually in transparent, plastic containers with a volume of 625 ml filled up to 460 ml, which were perforated on opposite sides to let water flow through. These containers were kept in a laminar flow tank with running low-copper water at 20 ± 1 C. The light conditions were 12:12 h. The snails were each fed one disc of lettuce, measuring ca. 19.6 cm 2 , daily. The individuals that were randomly selected to serve as sperm donors during the mating observations were kept 8 days in isolation to replenish the prostate gland and thus increase the motivation to mate as sperm donors. After 4 days their containers were replaced with clean ones. Egg laying is induced by a clean surface and/or clean water (Ter Maat et al., 1983). Inducing egg laying 4 days prior to mating observations avoids egg laying from interfering with mating. All donors were only used once as a mating partner in the experiment.

Treatment
We allowed the sperm donors to mate with only one of the following three different types of sperm recipients (also referred to as focal individuals, Fig. 1). The first type comprised recipients in which the male function was eliminated by removing the nervous connection between the prostate gland and the CNS. This was done by first anaesthetizing the snail with an injection of 2e3 ml of 50 mM MgCl 2 (needle gauge: 0.3 mm Â 13 mm) and ensuring that the individual was fully relaxed and extended from its shell (with its female gonopore visible). At this point the vas deferens, together with the thin branch of the penial nerve (NP1, De Boer et al., 1997) that runs alongside it, was carefully grasped with a thin #5 forceps, at the point where it runs very superficially under the thin body wall, and a small piece was cut . The loose ends of the vas deferens then retracted back into the body by themselves and the small hole in the body wall was left to heal by itself. The second type comprised recipients that were sham operated on, that is, they were also anaesthetized, and a similar-sized hole was made in the body wall right above the vas deferens, but the vas deferens and nerve branch were left intact. The third type of recipient comprised intact animals that served as the control group. Hence, the recipients that were operated on were feminized and could only mate in the female role (i.e. female recipients), while for the other two types both sexual roles were intact (i.e. hermaphroditic recipients). Each focal individual was marked with a small dot of nail polish on the shell for identification purposes during the subsequent mating observations. The focal individuals that were to become sperm recipients in the mating trial were kept in isolation for 4 days under the exact same conditions as the donor snails.

Behavioural Observations
For the mating observation we size-matched each focal individual with a donor, with no more than 0.3 cm difference in shell length between partners, because body size can also influence which mating role is chosen (Nakadera et al., 2015). Since the experiment required detailed observations per mating pair, only 9e12 focal individuals were observed per observation day (focal sampling, e.g. Altmann, 1974). To avoid confounding factors, on each day the same number of focal individuals was observed per treatment, that is, on each day three or four focal individuals of each treatment were observed. In total the interactions of 78 focal individuals with their mating partner were observed (26 operated, 27 sham-operated, 25 control).
During these observations the snails were placed in pairs in nonperforated transparent, plastic containers of 625 ml, filled up to 460 ml with water from the isolation tank. The room temperature was 20 ± 1 C. The pairs were continuously observed for 6 h; the observer took 1 min per pair (in turn) to note down the observed behaviour of the focal individual, then moved on to the next pair for 1 min, and so on. When all pairs were scored for 1 min, the cycle started again with the first pair (scan sampling, e.g. Altmann, 1974). Based on previous studies by Moussaoui et al. (2018) and Daupagne and Koene (2020), we decided to score the following behaviours: Locomotion, Crawl-out, Floating, Retraction, Biting and Shell shaking (see Table 1 for definitions). For all these behaviours we also scored whether the potential sperm donor was on top of the focal individual's shell or not. Finally, we recorded the time until insemination, that is, the time it took from the start of the experiment for the sperm donor to successfully inseminate the focal individual. Only pairs in which the focal individual was inseminated first were included in the analysis of time to insemination (N ¼ 55; 18 operated, 22 sham-operated, 15 control); pairs in which the focal individual performed the male role first or where no mating took place were left out (N ¼ 23: 8 operated, 5 sham-operated, 10 control).

Female Reproductive Success
After the mating observation, focal individuals were placed back into their tank in clean isolation containers. They were kept for 4 days and again fed 19.6 cm 2 lettuce disks daily. On the fourth day the egg masses were collected: they were carefully removed from the container wall using the flat side of a spatula and debris and excess water were removed by placing them on a paper towel before transfer onto a glass plate. The plates were turned upside down and placed on a flatbed digital scanner (Canoscan LiDE 700F Scanner) for scanning at 1200 dpi resolution. To count the eggs, brightness and contrast of the resulting images were adjusted (Van Iersel et al., 2014). After scanning, the masses were placed separately into 100 ml plastic tubes filled with low-copper water. The lid of each tube was perforated to allow for oxygen exchange with the air (and lids were not screwed shut). The water was refreshed every 4 days and on the 14th day after laying the hatched eggs were counted.

Operated Sham-operated Control
Hermaphroditic recipients Female recipient Figure 1. Experimental set-up. We created three different types of focal recipients. Operated treatment: recipients had their male function removed via a simple surgical lesioning of the nerve innervating the prostate gland (see Methods), that is, these were female recipients; sham-operated treatment: snails underwent a sham operation as a procedural control; control: intact snails. Hence, the latter two groups had both sexual roles intact (hermaphroditic recipients). experiment consisted of 7 observation days, we used these as a random factor in the mixed models. To test for differences in time until insemination, a linear mixed model (LMM) was used. The occurrence of avoidance behaviour was tested using generalized linear mixed models (GLMM, binomial distribution) with post hoc tests for pairwise comparisons using a DwasseSteeleCritchlowe-Fligner test. If the occurrence of an avoidance behaviour was significant, a GLMM was used to test for a relationship between the significant avoidance behaviour and time until insemination. Number of eggs laid and hatching success, after testing for normality of the data, were analysed using an LMM and a GLMM, respectively.

Ethical Note
The research adheres to the ASAB/ABS Guidelines for the Use of Animals in Research, the legal requirements of the Netherlands and all institutional guidelines. While the animal species used is not considered an Experimental Animal under European Law, we always did our utmost best to treat the animals very well (within the experimental constraints).

RESULTS
In total, 55 focal individuals were inseminated, and mating occurrence did not differ between treatments (c 2 ¼ 2.912, N ¼ 78, P ¼ 0.233). The 23 pairs that did not mate at all during the behavioural observations were excluded from the time until insemination analysis. The time it took the mating partner to inseminate the focal individual, that is, time until insemination, did not show a statistically significant difference between treatments (LMM: F 2, 13.36 ¼ 1.787, P ¼ 0.205; Appendix Fig. A1). For the analyses of the behaviours displayed by the sperm recipient (sometimes referred to in earlier literature as 'female copulants', e.g. in De Visser et al., 1994 andHoffer et al., 2010) the observation data of all 78 focal individuals were included. There was no significant difference between treatments in Locomotion and Crawl-out (GLMMs: Locomotion: c 2 ¼ 0.470, P ¼ 0.791; Crawl-out: c 2 ¼ 1.091, P ¼ 0.580; Fig. 2). Floating and Retraction could not be analysed because these behaviours occurred in too few pairs. The treatments did differ in the occurrence of Shell shaking (GLMM: c 2 ¼ 6.405, P ¼ 0.041; Fig. 2), but post hoc testing did not reveal a clear statistical difference between treatments; only sham-operated (hermaphroditic recipients) versus operated (female recipients) came close to statistical significance (DwasseSteeleCritchloweFligner test: W ¼ 3.227, P ¼ 0.058), with snails that were operated on showing a lower average occurrence. The occurrence of Biting differed significantly (GLMM: c 2 ¼ 13.524, P < 0.001; Fig. 2) and a post hoc test showed that snails that were operated on displayed this behaviour significantly less than snails in both other treatments (DwasseSteeleCritchloweFligner test: operated versus shamoperated: W ¼ 5.656, P < 0.001; operated versus control: W ¼ 4.978, P ¼ 0.001), while sham-operated and control snails did not differ from each other (W ¼ À0.658, P ¼ 0.879). However, the occurrence of Biting did not affect time until insemination (LMM: F 1, 2.88 ¼ 1.226, P ¼ 0.352).
We found no differences between treatments in our measures of female reproductive success. The mean number of eggs laid per snail in each treatment did not differ significantly between treatments (LMM: F 2, 40.03 ¼ 1.703, P ¼ 0.195). Also, hatching success after 14 days, which was calculated by dividing the number of hatched eggs by the total number of eggs deposited, did not differ significantly between treatments (GLMM: c 2 ¼ 0.265, P ¼ 0.876; Appendix Fig. A1).

DISCUSSION
This study revealed that surgical feminization of the pond snail L. stagnalis resulted in a decrease in one of the previously reported avoidance behaviours, namely Biting of the potential mating partner. Contrary to expectations based on previous work, this reduction in the occurrence of Biting did not lead to a reduction in the time until insemination of the focal individuals. Notably, the other prominent avoidance behaviour, Crawl-out, did not show any change and neither did egg laying or hatching success. The behavioural findings shed new light on the motivation that underlies avoidance behaviours in this simultaneous hermaphrodite.
Although feminized snails performed less Biting, the occurrence of the other female avoidance-like behaviours, Crawl-out, Shell shaking and Locomotion, did not differ significantly (but note that Shell shaking came close to statistical significance among the sham-operated and operated treatments). Thus, although Shell shaking has been reported to be used as a response to avoid unwanted inseminations in a smaller snail species (Physa acuta; Facon et al., 2006), this behaviour seems not to play a major role in the larger pond snail species studied here (confirming previous findings by Moussaoui et al., 2018). The latter study also reported Crawl-out as a potential avoidance behaviour, but this clearly did not differ between the female and hermaphroditic recipients of the current experiment. This implies that Crawl-out is a behaviour that is motivated through the female function of this simultaneous hermaphrodite. In contrast, the elimination of the male function did affect the occurrence of Biting behaviour, with the feminized snails (those that were operated on) showing Biting significantly less. This would mean that performing Biting behaviour is male role motivated or that at least the presence of also being able to perform the male role (for instance in the case of intertwined, hermaphroditic motivation) affects the occurrence of this behaviour.
While we saw a difference in one of the behaviours that has previously been reported to be implicated in potential avoidance of being inseminated (again), our results did not reveal a difference in the time until insemination between treatments. This may be explained by the space limitation in the observation container. As the snails were held in a relatively small space, successful avoidance of insemination may be less likely (Moussaoui et al., 2018). In addition, the set-up was also simplified because there was only one mating partner, which is an unlikely situation in both the field and our mass culture. Under more realistic circumstances, with many potential partners available, a slight discouragement of a mating partner may be sufficient. Moreover, our analysis was only based on the presence of the behaviour, and not on the frequency of Biting. Finally, the previously reported increase in time to insemination Table 1 Scored behaviours of the focal snails are listed with the definitions that we used during this study, based on Moussaoui et al., 2018 andDaupagne &. The behaviours are listed in the order as they appear in the Methods section

Behaviour Definition
Locomotion Individual crawls around on the container's surface Crawl-out Individual crawls above the water line via the container's wall and at least the head and/or part of the shell are positioned above the water line Floating Individual detaches its foot from the container surface and floats in the water or at the water surface (note that on the water surface they can locomote directionally) Retraction Individual remains stationary with its head hidden under its shell (note that this may also indicate the start of egg laying sequence) Biting Individual uses its radula (in an attempt) to rasp the shell or body of the mating partner that is mounted on the focals shell Shell shaking Individual raises its own shell and swings it from side to side while the mating partner is on top of the shell may be mainly due to the occurrence of Crawl-out, which did not differ in our study but did differ in the previous studies (Daupagne & Koene, 2020;Moussaoui et al., 2018). The reluctance to be inseminated by a mating partner may have several reasons. The study of Daupagne and Koene (2020) indicated that accessory gland proteins from the prostate gland and/or spermatozoa may cause the snails to be reluctant to mate in the female role, suggesting that receiving superfluous ejaculates is undesirable or disadvantageous and may reduce the individual's female fitness. In connection to this, the study by Nakadera, Swart et al. (2014) pointed out that accessory gland proteins reduce sperm transfer and paternity success of the receiving snails. This means that receiving ejaculates also impacts the male reproductive success. Hence, female and male reproductive success can be negatively affected by superfluous inseminations. Our finding that Biting occurred less when only the female function remained (after the male function was eliminated) may reflect that receiving an ejaculate was no longer a concern for these feminized hermaphrodites' male reproductive success. Thus, the regulation of Biting as a sexual avoidance behaviour may be induced by signals linked to the male side of this hermaphrodite's nervous system, such as the penial nerve and anterior lobe (Koene, 2010;2016;Di Cristo & Koene, 2017). Although several signal pathways influencing sexual behaviour in L. stagnalis have been distinguished, the signals responsible for avoidance behaviour remain to be examined (Koene, 2006(Koene, , 2010(Koene, , 2016Jarne et al., 2010;Di Cristo & Koene, 2017).
Clearly, further disentangling of sexual motivations for performing the male and/or female role and how these influence sexual avoidance behaviours within simultaneous hermaphrodites is needed. When considering Bateman's principle, one would expect selective mating to be motivated via the female role, because eggs are energetically more costly than spermatozoa (Bateman, 1948). However, when investigating simultaneous hermaphrodites, one should keep in mind that reproductive strategies and motivations of one sexual function always impact on the other sexual function. For example, when an individual avoids a mating interaction to be selective and increase its own egg production by avoiding an additional insemination, it will miss out on an opportunity to engage in a mating where the sexual roles are swapped (Koene & Ter Maat, 2007) and where it could donate sperm. As argued above, the current observations seem to suggest that Biting may be male role motivated (to avoid loss of paternity) while Crawl-out may be female role motivated (to avoid loss of egg production), even though they are both aimed at the same result, discouraging the potential mating partner from inseminating.
In this study we also tested the assumption that mating is energetically costly for hermaphroditic species, and that eliminating the male function frees up energy that can be reallocated towards female reproduction. This prediction is in line with previous research, such as that of Hoffer et al. (2010) where individuals were allowed to copulate for several days after which fecundity was assessed (i.e. number of eggs). That study only observed higher fecundity when both mating partners were feminized (see also , but not when individuals were experimentally restricted to one sexual role. In other words, reproductive resources were not reallocated when only one sexual role could be performed. Our findings in terms of egg number agree with this latter finding, but in our experimental set-up we only allowed for one observed mating interaction. Additionally, in our study hatching success seemed to be unaffected by only being able to reproduce via the female function. Whether hatching success is different in individuals that are only able to reproduce via the male function, i.e. do not receive ejaculate, remains to be demonstrated (but see Hoffer et al., 2017).
In conclusion, this research contributes to understanding how the presence of both sexual functions in simultaneous hermaphrodites affects individual motivations to mate in the male and/or female role and how this potentially affects reproductive success. Eliminating one of the sexual roles is a powerful method to test whether simultaneous hermaphrodites may change their reproductive strategies to optimize reproductive success. As shown here, feminized hermaphrodites seem to invest less in avoidance behaviour when the acceptance of ejaculates does not harm their paternity success but still seem to invest in avoidance behaviours related to maternal success. It would now be interesting to investigate how female role motivation is regulated mechanistically (i.e. in terms of willingness to receive an ejaculate, not in terms of egg laying; for the latter see e.g. Koene, 2010 for a review) and how this interacts at the CNS level with the male motivation neural network.
Disentangling these motivations provides a new and dynamic layer of complexity for sexual conflict processes in simultaneous hermaphrodites.

Declaration of Interest
Authors have no conflict of interest to declare.