Introduction
In many taxa, individuals use multiple cues to search for mates and engage in copulations . Along the years, several hypotheses have been put forward to explain the role of different cues in mate decision (see detailed classifications in : a) each cue may convey information about different mate qualities, the cues together increasing the accuracy of assessment, b) cues can be redundant but together improve discrimination, reducing errors associated to each cue, and/or c) cues can work differently in distinct environments and/or at dissimilar distances.
Recent work has been consolidating the importance of studying the role of multiple cues in varying and complex environments , under the assumption that individuals often experience fluctuations in their environment, such as changes in resource availability or in the density, quality and sex ratio of conspecific and heterospecific individuals. In such varying and complex environments, relying on multiple cues may contribute to an accurate and fast response by the receiver across environmental conditions . However, a disruption in the information transfer between signallers and receivers, via excess noise or a mismatch between cues, is also more likely in complex environments .
Mismatches among cues can occur when cues have different susceptibilities to changes in the environment, or when they persist for different periods of time . For example, mate quality can be perceived simultaneously via ephemeral cues, like many behavioural traits, and more permanent cues such as some morphological traits, with the former functioning as indicators of current condition thus changing with the environment, and the latter indicating lifetime performance and remaining virtually unchanged . That is the case of the field cricket (Gryllus campestris ) that uses both body size and chirp rate as cues for mate choice, typically giving priority to body size, the most permanent cue . In variable environments, however, fixed cues may become unreliable indicators of male quality , in which case the use of ephemeral cues may be favoured. A similar phenomenon is observed in thePieris rapae. During development, this butterfly relies on both temperature and photoperiod to predict conditions at adult emergence. Climate warming promotes the mismatch between these two cues, affecting temperature but not the photoperiod, which can result in a sub-optimal wing melanisation phenotype upon emergence, provided both cues continue to be used . The mismatches between cues can also occur because of sexual conflicts , with signallers emitting dishonest signals owning to diverging interests between sexes . Under all these scenarios, perceiving more than one cue can increase, rather than reduce, uncertainty about which decision to make.
While the costs of emitting multiple cues are extensively described , the costs for receivers are less well studied, especially when considering mismatches between cues. Besides the expected costs associated with increased energetic and cognitive effort needed to process more than one cue , costs for receivers are likely to be associated with missing opportunities of mating with a suitable mate or with attempting to mate with unsuitable mates . For instance, cues used to assess conspecific mate quality sometimes overlap with those used for species recognition, or kin recognition . This is the case inGryllus integer males for which the stimuli of heterospecific females overrides conspecific chemical cues during mating trials . To avoid losing suitable mates, one possibility is to evolve a lower mate acceptance threshold in environments with discordant cues , conceivably at the expense of accepting less desired mates . The same reduction in the acceptance threshold can be observed when the costs of assessing multiple cues are higher than the benefits of using them to only mate with suitable mates, in which case individuals can ignore the least reliable cues and accept a certain level of uncertainty . The opposite is expected if the goal is to prevent matings conferring little to no reproductive success. Therefore, the optimal use of cues and corresponding behaviour should depend on the balance between the costs of acceptance and rejection errors .
So far, the studies exploring the response of individuals to mismatches between cues in the context of mate choice have largely disregarded cases in which males are receivers and females are signallers of cues. Addressing this issue can be particularly relevant in species with first male sperm precedence, where female mating status discrimination is essential for male mating success. Indeed, under this pattern of sperm precedence, mating with mated females provides low, if any, fertilization opportunities, whereas mating with virgin females strongly contributes to reproductive success . Accordingly, it has been shown that males of these species have evolved the ability to discriminate female mating status and modulate their reproductive behaviour based on the cues presented by females . Yet, the behaviour of males when cues provide discordant information concerning the female mating status, as well as the associated costs thereof, remain largely unknown.
To fill this gap, we observed the mating behaviour of male and female two-spotted spider mites (Tetranychus urticae) in environments with information concerning the female mating status coming from two sources, the female itself and the cues it leaves in the substrate. Spider mites have first male sperm precedence and, accordingly, males prefer to mate with virgins, basing their decision upon volatile and substrate cues . Furthermore, matings with virgin females take less time to start, are longer, and induce more survival costs in males than matings with mated females . All this suggests that male reproductive investment in matings with virgins and with mated females is not the same. However, matings involving mated females are frequently observed in laboratory populations , despite often leading to lower fecundity . This suggests that discrimination in this species is not perfect and may depend on the composition of cues present in the environment. Spider mite populations occur in variable environments, as they colonize seasonal resources such as agricultural crops . Moreover, they disperse among patches after a variable number of generations in the same patch, following a subdivided haystack population structure . This results in a scenario of cyclic waves of virgin and mated females across time within the same plant, which fosters the conditions for a mismatch between the different cues signalling female mating status. Here, we tested the consequences of cue mismatch within this context.