Abstract In many species, adults exploit sensory information experienced in their natal habitat when searching for resources. This behavioral plasticity may help animals to establish themselves in new habitats by quickly locating suitable resources and avoiding unsuitable resources in complex environments. However, the processes guiding positive or negative natal habitat preference induction (NHPI) remain poorly understood. In the polyphagous moth Spodoptera littoralis, earlier studies have shown that female innate host-plant preference is modulated by larval feeding experience. In this context, the aim of this study was to investigate how variability in food quality associated with habitat olfactory cues can modulate NHPI in this species. We found that larvae showed appetitive or aversive responses to the experienced plant olfactory cues based on their values as predictors of food quality. Furthermore, larval exposure to host-plant olfactory cues alone induced oviposition preference for these plants in adult females, but only when the females had been feeding on high-quality food as larvae. Females reared on poor quality food retained their innate plant oviposition preference as adults. These results show that NHPI in S. littoralis is context-dependent and based on food quality with which olfactory cues are associated. They also suggest that larval experience to plant olfactory cues alone is sufficient to modulate the adult host-plant preference. Finally, this study suggests that polyphagous insects with particular innate plant preferences may only show phenotypic plasticity in this trait when the fitness benefits are high. INTRODUCTION One constant of nature is its perpetual change and the main challenge of an organism is to maximize its fitness under such variable environmental conditions. Behavioral plasticity, i.e. the ability of one genotype to produce multiple behavioral phenotypes under different environmental conditions, is one way to cope with changing and unpredictable conditions in nature and to adapt to new environments (Dukas 2008; Snell-Rood 2013). Such phenotypic plasticity is considered to be favorable in highly variable environments where no single behavioral phenotype is consistently optimal (Price et al. 2003; Ghalambor et al. 2007; Mery and Burns 2010). The benefits of behavioral plasticity are especially important in environments that vary between generations, but remain relatively stable within generations (Snell-Rood 2013). Some animals have evolved the ability to detect environmental cues and store reliable information from certain life experiences to fine-tune their behavior and better adapt to the present conditions (Mery and Burns 2010). Behavioral plasticity, based on prior experiences, allows for rapidly induced responses to a specific environment. Experiences acquired at different stages of an individual’s life can represent a valuable source of information about the current state of the environment. In a broad range of animal species, sensory cues experienced early in life are used in the adult stage to identify resources with traits similar to those encountered in their natal habitat (Davis and Stamps 2004). This phenomenon has been described as natal habitat preference induction (NHPI) (Davis and Stamps 2004) and has primarily been supported by studies on insects (Davis 2008), yet it occurs in mammals and birds as well (Selonen et al. 2007, Piper et al. 2013). In insects, the transfer of information guiding host-plant preference over a full metamorphosis has also been called Hopkins’ host selection principle (HHSP; Hopkins 1917; Barron 2001). NHPI has also been suggested to have large influence on behaviors that are important for many central ecological research questions, for example, dispersal and metapopulation dynamics (Benard and McCauley 2008), host-range expansions (Zhang et al. 2007), and ecological divergence and speciation events (König et al. 2014). It has been proposed that past experiences from the natal environment can increase the performance of individuals in the new habitat when the new situation is similar to what it was previously experienced by the individual (Davis and Stamps 2004; Stamps and Davis 2006). For instance, individuals may be able to better utilize certain resources present in the new habitat or reduce the costs of decision-making in complex situations by using previous information to assess the quality of the new habitats (Carrasco et al. 2015). The behavioral outcome of the preference induction may thus depend on whether the experience is positively or negatively reinforced. NHPI is usually referred as a positive induction of preference for cues from the natal environment and would lead to associational effects between sensory cues in the natal environment and a positive experience (Davis 2008). Furthermore, theoretical models predict that positive induction should be stronger than negative induction in changing preferences and driving phenotypic plasticity effects (Stamps et al. 2009). However, negative preference induction could still occur from natal habitat experience leading to a reduced attraction towards habitats presenting similar cues as the natal habitat. The habitat of herbivorous insects can show large spatial and temporal variability in relation to availability and density of host plants between seasons and herbivore generations. However, as they undergo rapid development, the chance is high that, within a generation, adult individuals experience a similar environment as that experienced in the juvenile stage. For herbivorous insects, and particularly generalist species, the selection of a suitable host plant can be a critical and complicated task when quality in relation to larval fitness differs among plants. They may have to locate and sample many different potential host plants in their environment before a plant is accepted for feeding or oviposition. Furthermore, generalist species may have neural limitations that reduce their ability to detect or process the relevant plant cues that are available (Bernays 2001). Information from earlier host-plant experiences could thus be very beneficial to make fast and accurate decisions when selecting a host plant. In the last 2 decades, a number of studies have demonstrated NHPI and HHSP in herbivorous insects (Anderson and Anton 2014). However, there are also studies that have failed to demonstrate the induction of preference for cues from natal habitats or host plants (Barron 2001; Janz et al. 2009). In fact, very little is still known about the processes guiding NHPI in insects using experience during host-plant choice. There are examples showing that this information transfer is based on olfactory cues (Anderson and Anton 2014). However, it is still not known how natural odor blends from host plants are involved in NHPI and if this complex signal can be transferred over the metamorphosis and used by the adult insect when selecting host plants. Further investigations are needed to disentangle what experiences induce preference shift in the adult, what sensory cues are used for the induction of preference for a specific plant and how information is transferred. The moth Spodoptera littoralis is a highly polyphagous species that can utilize many different host plants (Pogue 2002). Previous studies have shown that this moth has an innate preference hierarchy between different potential host plants and that larval feeding experience is transferred over metamorphosis and affects both male and female adult host-plant preference (Thöming et al. 2013; Proffit et al. 2015). In these studies, an increased adult preference was found for the experienced plant during larval development in 4 out of 5 plant species, but not for an innately less preferred plant. This result suggested that NHPI mechanisms are involved in the modulation of host-plant preference and also that such plasticity in plant preference is selective and could be associated with the estimated quality of the larval food plant. It is possible that only in situations when the quality of the larval food sustains a good larval development would a transfer of information occur that can be used by the adult moth. Thus, a positive larval feeding experience would affect adult host-plant choice, whereas a negative feeding experience could induce a negative induction of preference or alternatively not promote a plastic response to host-plant cues. These experiments also indicated that olfactory cues could be important for host-plant choice and that these cues can be used to identify host plants and to predict plant quality (Thöming et al. 2013). Furthermore, larval feeding experience can induce both appetitive and aversive responses in subsequent larval feeding occasions (Salloum et al. 2011). Thus, this species provides us with an excellent model system to test phenotypic plasticity based on positive and negative experiences on NHPI and associative effects based on complex olfactory cues directly involved in host-plant selection in herbivorous insects. In this study, we aimed to test how larval olfactory experience, associated with a positive or negative feeding experience, affected both larval and adult behavioral decision-making during host-plant choice in S. littoralis. During the experiments, we provided external olfactory cues from host plants in conjunction with larval feeding on artificial diets of different qualities as food for larval development. It allowed us to study if the transfer of information is selective and dependent on the valence of the larval experience. It also allowed us to isolate olfactory sensory input from any additional cues associated with the ingestion of the food substrate. The aim was to assess 1) if food quality could modulate the induction of preference, 2) if different aspects of food quality, such as nutritional content or the presence of secondary metabolites, mediated induction of preference in different ways, and 3) if olfactory host-plant cues experienced during the larval stage, isolated from dietary experience, could be transferred over metamorphosis and drive adult phenotypic plasticity through NHPI. MATERIALS AND METHODS Insects The rearing strain of S. littoralis was founded from moths collected in the Alexandria region in Egypt in 2008 and it has been supplemented every year with moths collected in the wild in Egypt. Plants We used 4 plant species known to elicit different degrees of innate preferences in ovipositing females of S. littoralis (Thöming et al. 2013); ranked from the highest to the lowest preference: Cowpea (Vigna unguiculata subsp. unguiculata, Fabaceae), Egyptian clover (Trifolium alexandrinum, Fabaceae), cotton (Gossypium hirsutum, v. Delta Pineland 90, Malvaceae), and cabbage (Brassica oleracea v. capitata, Brassicaceae). They were all grown from seeds for 5–6 weeks in a greenhouse (25 ± 2 °C, 70 ± 2 % RH and 16:8 h light:dark) until they were used in experiments. Plants were potted in 1.5 L pots in a commercial soil (Kronmull, Weibull Trädgård AB, Hammenhög, Sweden). All plants were used at the nonflowering stage. Diets Three artificial diets differing in nutritional quality were used: a control diet (described below), a diet with lowered nutritional content by reducing the amount of protein and fat (hereinafter referred to as low-protein diet), and a diet with added caffeine, a plant secondary compound known to be a feeding deterrent for insects (Shields et al. 2008; hereinafter, caffeine-supplemented diet). The control diet was based on potatoes with added wheat-germ and dried-yeast flakes as main protein sources (Hinks and Byers 1976). The low-protein diet was made by replacing two-thirds of the wheat-germ and dried-yeast flakes by potato starch and the caffeine-supplemented diet was made by adding 2% of the dry food weight in caffeine powder (ReagentPlus® 99% purity, Sigma-Aldrich, USA) to control diet (see detailed diet recipes in Supplemental Material). Plant odor exposure and oviposition preference In the first experiment, larvae were only reared on the control diet in plastic boxes (18 × 24 × 7 cm) with windows covered with stainless steel wired mesh (opening: 0.1 mm, thickness: 0.0063 mm) at each side and on the top until pupation in a greenhouse (25 ± 5 °C, 50–70 % RH and 16:8 h light:dark). These windows allowed exposure to plant volatiles while avoiding any contact between the larvae and plant material. The rearing boxes containing the larvae were placed in polyester mesh cages (47 × 47 × 138 cm). In these cages, larvae were either exposed to volatiles from 3 cotton plants, 3 cabbage plants, or they were kept without plants serving as control (no plant odor exposure). Cages were isolated from each other with plastic curtains to prevent the transmission of plant odors between cages. To exclude the influence of early adult experience, pupae were removed from the cages used during larval development. The collected pupae were sexed, and the female pupae were kept separately in a plant odor free climate chamber (25 ± 2 °C, 70 ± 2% RH, 15:9 h light:dark) until adult emergence. Newly emerged females were isolated daily to control their age, provided with honey water (1:1) and then mated 24 h after emergence with males from the main rearing fed on artificial diet. Only 2–4 day old, mated females were used during the experiments. Adult females were tested individually in 2-choice oviposition experiments where those exposed to cotton plants were given the choice between 1) a cotton and a cowpea plant or 2) a cotton and a clover plant, whereas those exposed to cabbage plants were given a choice between 3) a cotton and a cabbage plant. In all 3 experiments, a control using females with no exposure to plant odor was performed. After 3 days, the egg batches laid on each host plant were removed and weighed individually. Number of replicates in each plant combination larval-treatment group varied between 24 and 29 tested individuals. Diet quality and larval performance To test the effect of diet quality on larval performance, larval developmental time, pupal mass (1 day after pupation), and survival were recorded for insects reared on the 3 different diets: control diet (n = 697), caffeine-supplemented diet (n = 334), and low-protein diet (n = 251). Effect of diet quality and larval preference To test the combined effect of diet quality and larval olfactory experience on larval host-plant preference, larvae were reared under the same conditions and using the same method as in the plant exposure experiments described above. In a 3 × 3 factorial experiment (Figure 1), larvae were first arbitrarily divided into 9 groups of 40 individuals. Three cohorts of larvae were reared on each of the 3 diets: control, low-protein, or caffeine diet. Then, one group of larvae from each of the diets was placed in cages (47 × 47 × 138 cm) and exposed to the odor of either 3 cowpea plants, 3 cotton plants, or no plants throughout their development. Ten third-instar larvae were taken from each of the 9 different treatments and tested singly for host-plant preference in a 2-choice olfactometer. These larvae were not re-introduced into the rearing after the experiment. The remaining larvae were reared to the adult stage and 2–4 day old females were used for the oviposition preference experiments (see below). This factorial experiment was repeated 8 times. Figure 1 View largeDownload slide Experimental design used to test the combined effect of diet quality and plant odor exposure on larvae and adult host-plant choice. Larvae were reared in plastic boxes with windows covered with stainless steel wired mesh to prevent larvae to escape but at the same time allow exposure to plant volatiles. The rearing boxes containing the larvae were placed in polyester mesh cages. In these cages, larvae were either exposed to volatiles from 3 cotton plants, 3 cowpea plants, or they were kept without plants serving as control (no plant odor exposure). Larvae were either fed with a control diet, a diet with lowered nutritional content or a diet with added caffeine. Figure 1 View largeDownload slide Experimental design used to test the combined effect of diet quality and plant odor exposure on larvae and adult host-plant choice. Larvae were reared in plastic boxes with windows covered with stainless steel wired mesh to prevent larvae to escape but at the same time allow exposure to plant volatiles. The rearing boxes containing the larvae were placed in polyester mesh cages. In these cages, larvae were either exposed to volatiles from 3 cotton plants, 3 cowpea plants, or they were kept without plants serving as control (no plant odor exposure). Larvae were either fed with a control diet, a diet with lowered nutritional content or a diet with added caffeine. For each of the 9 treatments described above, 8 replicates of 10 larvae each were tested for their plant preference for cotton and cowpea using a 2-choice olfactometer (28 × 20 × 5.5 cm; see experimental setup in Carlsson et al. (1999)). The initial ambient air was pumped through a tube containing activated charcoal into polyethylene cooking bags (Meny, Toppits, 45 × 55 cm) containing one of the odor sources (cotton plant or cowpea plant). Each odor source was connected with a teflon tube to one of the short side of the olfactometer. The odor was drawn into the olfactometer (airflow: 0.4 L/min) creating 2 flows from each side that meet in the middle of the olfactometer. The air was then sucked out through a hole in the lid in the center of the olfactometer by a tube connected to a second pump. Before every experiment, flow meters were used to adjust and control the airflows and the olfactometer was rinsed with ethanol (70%) and aired for approximately 5 min. At the start of the experiments, larvae were individually placed on the midline of the olfactometer and exposed to the converging airflows. A larva was considered to have made a choice when reaching within 1 cm from the nylon mesh walls at the peripheral end of the olfactometer. Larvae that did not make a choice within 5 min were not further analyzed. Depending on the treatment, from 13 to 20 of the tested larvae (n = 80 per treatment) did not make a choice. Diet quality and oviposition preference Inexperienced females (control diet, n = 26; low-protein diet, n = 19; caffeine-supplemented diet, n = 20), cotton-experienced females (control diet, n = 28; low-protein diet, n = 16; caffeine-supplemented diet, n = 19), and cowpea-experienced females (control diet, n = 25; low-protein diet, n = 23; caffeine-supplemented diet, n = 22) were tested for their oviposition preference when given the choice between a cotton and a cowpea plant. These experiments were performed in a greenhouse (25 ± 5 °C, 50–70% RH, 15:9 h light:dark) in wooden cages (120 × 60 × 80 cm) with sidewalls of wire gauze (mesh size: 2.5 mm). One cotton plant and one cowpea plant were positioned at opposite sides of the cage. One male from the external rearing and one treated female were mated and placed in each cage. After 3 days, the egg batches laid on each host plant were removed and weighed individually. Statistical analyses The statistical analyses were performed using R software (R Development Core Team 2015). The age at pupation was compared among diets using a generalized linear model (GLM, R function glm) with a Poisson error distribution, the pupal weight was compared using a GLM with a Gaussian error distribution, and the larval survival rate was compared using a GLM with a binomial error distribution. Post hoc pairwise comparisons were made using Tukey’s HSD test (R function glht from the R package multcomp; (Hothorn et al. 2008)). Larval host-plant preference indices (LPIs) were evaluated based on the total choice per replicate of the larvae for both plants and calculated as follows: LPI=(cotton choice−cowpea choice)total choice The adult female plant preference indices (FPIs) were based on the total weight of eggs laid on both plants by each female and calculated as follows: FPI=(egg mass on cotton−egg mass on cowpea)total mass eggs For both larvae and adult females, the indices range from 1 (absolute preference for plant A) to −1 (absolute preference for plant B), with 0 meaning no preference. These indices were used for data illustration. For the statistical analyses, we used the choice of larvae for either plant species (binary variable) and the proportion of egg mass laid on either plant by females. A generalized linear model (GLM, R function glm) was fitted to the data and analyzed using the Bernoulli error distribution and logit link function for larval choice data and using binomial error distribution with the logit link function for adult choice data. Post hoc pairwise comparisons between plant-pair combinations were made using the Tukey’s HSD test (R function glht from the R package multcomp; (Hothorn et al. 2008)). RESULTS Effect of larval odor exposure on adult female host-plant choice The results of the odor exposure experiments showed that exposure to plant volatiles during the larval stage had a significant effect on host-plant choice of ovipositing females of S. littoralis. When exposed to cotton plants during the larval stage, females subsequently showed a shift in their egg laying preference towards cotton over cowpea (GLM, χ2 = 3.9, df = 1, P = 0.03; Figure 2a) or clover plants (GLM, χ2 = 13.8, df = 1, P < 0.001; Figure 2b) compared with the nonexposed control females. In the same way, females exposed to cabbage plant odor as larvae subsequently shifted their egg laying preference towards cabbage over cotton (GLM, χ2 = 11.3, df = 1, P < 0.001; Figure 2c) compared to nonexposed females. Figure 2 View largeDownload slide Effects of plant odor exposure during the larval stage (squared = cotton exposure, broken lines = cabbage exposure, grey = no odor exposure) on adult female plant oviposition preference (mean ± SE of the preference index) when given the choice between cotton and cowpea (a), cotton and clover (b), and cotton and cabbage (c). Asterisks indicate significant differences between treatments (GLM, Tukey’s HSD test; *P < 0.05, ***P < 0.001). Figure 2 View largeDownload slide Effects of plant odor exposure during the larval stage (squared = cotton exposure, broken lines = cabbage exposure, grey = no odor exposure) on adult female plant oviposition preference (mean ± SE of the preference index) when given the choice between cotton and cowpea (a), cotton and clover (b), and cotton and cabbage (c). Asterisks indicate significant differences between treatments (GLM, Tukey’s HSD test; *P < 0.05, ***P < 0.001). Effect of diets on larval performance Diet had a significant effect on larval developmental time (GLM, χ2 = 648.33, df = 2, P < 0.001; Figure 3a), pupal mass (GLM, χ2 = 11.8, df = 2, P < 0.001; Figure 3b), and larval survival (GLM, χ2 = 327.35, df = 2, P < 0.001; Figure 3c). Larval performance in terms of larval development and pupal mass was highest on control diet, intermediate on caffeine-supplemented diet, and lowest on low-protein diet (Figure 3). Compared with the performance on control diet, larval development was extended by 12% and 14% and pupal weight was reduced by 10% and 22% on caffeine-supplemented and low-protein diet, respectively. Survival of the larvae was also highest when the larvae were fed on control diet and more than 25% higher than on the other 2 diets. Figure 3 View largeDownload slide Effect of larval diets (control, caffeine-supplemented and low-protein diet) on larval developmental time (= age at pupation) (a), pupal weight (b), and larval survival (c). Box plots show the mean (black square), median (white line), and 25–75% percentiles. Whiskers show all data excluding outliers. Outliers (circles) are values being more than 1 time box length from upper and lower edge of respective box. Letters indicate significant differences between treatments (GLM, Tukey’s HSD test, P < 0.05). Figure 3 View largeDownload slide Effect of larval diets (control, caffeine-supplemented and low-protein diet) on larval developmental time (= age at pupation) (a), pupal weight (b), and larval survival (c). Box plots show the mean (black square), median (white line), and 25–75% percentiles. Whiskers show all data excluding outliers. Outliers (circles) are values being more than 1 time box length from upper and lower edge of respective box. Letters indicate significant differences between treatments (GLM, Tukey’s HSD test, P < 0.05). Effect of diet and odor exposure association on larval choice We found a significant interaction between odor exposure and diets on larval plant preference (GLM, χ2 = 25.19, df = 4, P < 0.001; Figure 4). Diet did not influence the preference between cotton and cowpea for third-instar larvae with no prior plant odor experience (GLM, χ2 = 0.39, df = 2, P = 0.81; Figure 4, gray bars). However, larvae fed on control diet and exposed to plant odors showed preferences shift towards the odors of the plant to which they had been exposed previously (Z = −3.2, P < 0.01; Figure 4). In contrast, a shift of preference towards the nonexperienced odor occurred when larvae had been fed on the diets with inferior quality (caffeine-supplemented diet, Z = 3.2, P < 0.01; low-protein diet, Z = 2.18, P < 0.05; Figure 4). Figure 4 View largeDownload slide Combined effects of plant odor exposure (gray: control, squared: cotton, dotted lines: cowpea) and diet (control diet, low-protein, and caffeine-supplemented diet) on larval and female plant preference (mean ± SE of the preference index) when given the choice between a cotton (+1) and a cowpea plant (−1). Letters indicate significant differences between treatments (GLM, Tukey’s HSD test; P < 0.05). Figure 4 View largeDownload slide Combined effects of plant odor exposure (gray: control, squared: cotton, dotted lines: cowpea) and diet (control diet, low-protein, and caffeine-supplemented diet) on larval and female plant preference (mean ± SE of the preference index) when given the choice between a cotton (+1) and a cowpea plant (−1). Letters indicate significant differences between treatments (GLM, Tukey’s HSD test; P < 0.05). Effect of diet and odor exposure on oviposition Diet did not modify egg laying (oviposition) preference between cotton and cowpea for females with no prior plant odor experience during larval development (GLM, χ2 = 0.4, df = 2, P = 0.81; Figure 4, gray bars). Females fed on control diet and exposed to plant odors showed preferences shifted toward the odors of plant to which they had been exposed during larval development (Z = 2.6, P < 0.05; Figure 4). In contrast, no shift of oviposition preference due to odor exposure was observed when females were fed on caffeine-supplemented diet (GLM, χ2 = 0.68, df = 2, P = 0.71; Figure 4) or on low-protein diet during the larval stage (GLM, χ2 = 0.14, df = 2, P = 0.93; Figure 4). DISCUSSION Our previous studies have shown that S. littoralis has an innate preference hierarchy between different potential host plants and that larval feeding on a specific host plant normally increases the preference for this plant to become the most preferred by the adult (Anderson et al. 2013; Thöming et al. 2013; Proffit et al. 2015). The present work expands on these studies and provides evidence for NHPI based on the transfer to the adult of olfactory information from the natal habitat, modulated by the quality of the food experienced by the larvae. We found here that NHPI based on larval experience to host-plant cues in the moth S. littoralis is selective and context-dependent. When larvae were fed on a high-quality artificial diet, combined with exposure to olfactory cues from a host plant, an increased preference was found for the experienced plant odor in both larvae and ovipositing adult females. For diets of lower quality, where both the caffeine-supplemented diet and low-protein diet reduced pupal weight and increased mortality, avoidance behavior for the experienced plant odor was found in larvae. However, no preference change was found in the adult by larval exposure to host-plant odor when reared on these low-quality diets. Thus, the change in adult preference based on larval olfactory experience was limited to the high-quality diet. The context-dependent change of adult preference is also supported by the responses of adult S. littoralis to the nonpreferred host-plant cabbage and cabbage odor. In the current study, an increased oviposition preference for cabbage was induced when larvae experienced cabbage odor while feeding on a high-quality artificial diet. In contrast, we have previously found that when larvae were allowed to feed directly on cabbage leaves, no increased preference for cabbage was found for adult moths (Thöming et al. 2013). Thus, larval feeding on cabbage does not induce a preference for this plant in adults, whereas cabbage volatiles presented in another feeding context can induce a preference for this plant. Cabbage plants contain secondary plant compounds, for example, glucosinolates, that are known to negatively affect the development of generalist herbivores (Gols et al. 2008; Kos et al. 2012). The presence of these secondary compounds could explain that a feeding experience on cabbage does not induce a preference for this plant, just as we in this study found that adding caffeine to a high-quality diet affected the preference induction. This shows that, in the case of cabbage, the odor is not always repellent, but it is rather the context in which the odor is perceived and to what experience the odor is associated that decides its valence. There are several examples from herbivorous insects that show a positive habitat preference induction (Anderson and Anton 2014), whereas, to our knowledge, a negative induction of habitat preference has only been shown in the specialist moth Choristoneura fumiferana (Mader et al. 2012). Another study provided evidence for the memory retention of aversion through metamorphosis in a moth. However, the authors used a system where larvae were conditioned to odors while at the same time being exposed to an electric shock (Blackiston et al. 2008). The repeated training resulted in conditioned avoidance of the shock-associated odor in larvae and in adult females after emergence. Although the latter experiments reveal the capability of some species for conditioned avoidance, this behavior has never been demonstrated using ecologically relevant reinforcing stimuli. Several theories have been suggested, which could explain why negative NHPI is not common in nature. One reason could be that 1) individuals that performed poorly in their natal habitat would be less likely to survive and disperse efficiently (Stamps and Davis 2006). Another reason could be that 2) insects from poor environments show lower selectivity (Stamps 2006). However, it may not be the case for S. littoralis, where we found that insects reared on both high- and low-quality diets were selective in their oviposition choice, but they differed in the way they used the larval experience. Females from high-quality diets diverged from the innate preference hierarchy and selected the plant that they had experienced as larvae, whereas females from low-quality diets still relied on the innate hierarchy between the host plants. Furthermore, it has been predicted that 3) NHPI is only advantageous if natal experience induces preference for resources that are of higher quality than dispersers would have assumed in the absence of experience (Stamps et al. 2009). The benefits from using information acquired in the natal habitat must outweigh the associated costs (Webster et al. 2013). For highly polyphagous insects, like S. littoralis, negative experiences may not benefit decision-making during host-plant choice to the same degree as positive experiences. A positive experience with a specific host plant would increase salience for cues from that plant and could facilitate searching for this plant, whereas avoidance of one specific host plant may not allow host-plant choice to be more efficient. Finally, it has been suggested that 4) negative preference induction should also be less common in animals already exhibiting a strong innate preference for different hosts or habitats (Webster et al. 2013). Adult females may rely on their innate plant preference hierarchy instead, rather than showing behavioral avoidance of a specific host plant and allocating energy into a costly investment in aversion memory (Burns et al. 2011; Plaçais and Preat 2013). For example, the black bean aphid Aphis fabae has been shown to be able to associate plant odors with visual cues indicating accessibility of the food source (Webster et al. 2013). Interestingly, the aphids relied on the learned information for host-plant choice only when the perceived cost of habitat assessment was low. Otherwise, they relied on their innate plant preferences, suggesting an ability to make use of either learned or innate behavioral responses depending on the environmental context (Webster et al. 2013), which also could be the case for S. littoralis. In this study, we also showed that olfactory cues alone are sufficient for information transfer from larvae to adult as we in our experiments disassociated the experience to olfactory cues from other sensory cues from the host plant. These results confirm and extend on our previous studies that show that olfactory cues are important for host-plant choice decision in adult S. littoralis (Thöming et al. 2013; Zakir et al. 2013). Our data are also consistent with associative processes between diet quality and behavioral response to plant olfactory cues. For larval attraction to plant odor, both positive and negative associations between the experienced plant odor and diet quality was found, whereas only positive associations were found for adult oviposition behavior. The olfactory blends of the host plants of S. littoralis used in the experiments in this study show overlap in the emitted compounds, but that each plant has a unique blend of compounds (Conchou et al. 2017) and an image of this blend needs to be transferred over the full metamorphosis to affect adult preference. For HHSP, 2 main mechanisms have been proposed to explain this information transfer. Firstly, learned olfactory cues experienced by the larvae are transferred to the adult through memory retention (Tully et al. 1994; Blackiston et al. 2008), where the olfactory information would be stored in the mushroom bodies, as this structure is considered to be involved in memory formation within the brain (Heisenberg 2003). Although larval mushroom bodies are extensively rearranged during metamorphosis, there is neurobiological evidence showing transfer to the adult of specific mushroom body neural components through metamorphosis in insects (Armstrong et al. 1998, Ray 1999). A second suggested mechanism is “chemical legacy” that hypothesizes that host-plant cues can remain, inside or outside the pupa, and prime a preference for these odors at adult emergence (Corbet 1985). Even though our results show that the adult preference induction occurs even when odor stimuli are disassociated from the larval food, our results do not provide clear support for either of these mechanisms. Further studies are needed to show how the olfactory information is transferred and affects adult behavior. To conclude, our study demonstrates that 1) NHPI is context-dependent and only occurs when the larvae are fed a high-quality diet, and that 2) larval olfactory experience is sufficient to induce NHPI in adult moths. These findings provide new evidence of an induction of natal habitat preference in phytophagous insects and should stimulate future studies combining behavioral ecology and sensory biology to investigate the neurophysiological bases underlying this phenomenon. SUPPLEMENTARY MATERIAL Supplementary data are available at Behavioral Ecology online. FUNDING This work was supported by the Linnaeus grant “Insect Chemical Ecology, Ethology and Evolution”, funded by the Swedish Research Council for Envirnonment, Agricultural Sciences and Spatial Planning (Formas) and by the Max Planck Society. We would like to thank Elisabeth Marling for the rearing of the experimental moths and Thomas Svensson for assistance with experimental assays. We also thank Kristina Karlsson Green and Rieta Gols for valuable comments to earlier versions of the manuscript. Data accessibility: Analyses reported in this article can be reproduced using the data provided by Lhomme et al. (2017). REFERENCES Anderson P, Anton S. 2014. Experience-based modulation of behavioural responses to plant volatiles and other sensory cues in insect herbivores. Plant Cell Environ . 37: 1826– 1835. Google Scholar CrossRef Search ADS PubMed Anderson P, Sadek MM, Larsson M, Hansson BS, Thöming G. 2013. Larval host plant experience modulates both mate finding and oviposition choice in a moth. Animal Behav . 85: 1169– 1175. Google Scholar CrossRef Search ADS Armstrong JD, de Belle JS, Wang Z, Kaiser K. 1998. Metamorphosis of the mushroom bodies; large-scale rearrangements of the neural substrates for associative learning and memory in Drosophila. Learn Mem . 5: 102– 114. Google Scholar PubMed Barron AB. 2001. The life and death of Hopkins’ host selection principle. J Insect Behav . 14: 725– 737. Google Scholar CrossRef Search ADS Benard MF, McCauley SJ. 2008. Integrating across life-history stages: consequences of natal habitat effects on dispersal. Am Nat . 171: 553– 567. Google Scholar PubMed Bernays EA. 2001. Neural limitations in phytophagous insects: implications for diet breadth and evolution of host affiliation. Annu Rev Entomol . 46: 703– 727. Google Scholar CrossRef Search ADS PubMed Blackiston DJ, Casey ES, Weiss MR. 2008. Retention of memory through metamorphosis: Can a moth remember what it learned as a caterpillar? PLoS One 3: e1736. Google Scholar CrossRef Search ADS PubMed Burns JG, Foucaud J, Mery F. 2011. Costs of memory: lessons from “mini” brains. Proceedings. Biol Sci R Soc . 278: 923– 929. Google Scholar CrossRef Search ADS Carlsson MA, Anderson P, Hartlieb E, Hansson BS. 1999. Experience-dependent modification of orientational response to olfactory cues in larvae of Spodoptera littoralis. J Chem Ecol . 25: 2445– 2454. Google Scholar CrossRef Search ADS Carrasco D, Larsson MC, Anderson P. 2015. Insect host plant selection in complex environments. Curr Opin Insect Sci . 8: 1– 7. Google Scholar CrossRef Search ADS Conchou L, Anderson P, Birgersson G. 2017. Host plant species differentiation in a Polyphagous Moth: Olfaction is enough. J Chem Ecol . 43: 794– 805. Google Scholar CrossRef Search ADS PubMed Corbet SA. 1985. Insect chemosensory responses: a chemical legacy hypothesis. Ecol Entomol . 10: 143– 153. Google Scholar CrossRef Search ADS Davis JM. 2008. Patterns of variation in the influence of natal experience on habitat choice. Q Rev Biol . 83: 363– 380. Google Scholar CrossRef Search ADS PubMed Davis JM, Stamps JA. 2004. The effect of natal experience on habitat preferences. Trends Ecol Evol . 19: 411– 416. Google Scholar CrossRef Search ADS PubMed Dukas R. 2008. Evolutionary biology of insect learning. Annu Rev Entomol . 53: 145– 160. Google Scholar CrossRef Search ADS PubMed Ghalambor CK, McKay JK, Carroll SP, Reznick DN. 2007. Adaptive versus non-adaptive phenotypic plasticity and the potential for contemporary adaptation in new environments. Funct Ecol . 21: 394– 407. Google Scholar CrossRef Search ADS Gols R, Wagenaar R, Bukovinszky T, van Dam NM, Dicke M, Bullock JM, Harvey JA. 2008. Genetic variation in defense chemistry in wild cabbages affects herbivores and their endoparasitoids. Ecology . 89: 1616– 1626. Google Scholar CrossRef Search ADS PubMed Heisenberg M. 2003. Mushroom body memoir: from maps to models. Nat Rev Neurosci . 4: 266– 275. Google Scholar CrossRef Search ADS PubMed Hinks CF, Byers JR. 1976. Biosystematics of the genus Euxoa (Lepidoptera: Noctuidae). V. Rearing procedures, and life cycles of 36 species. Can Entomol . 108: 1345– 1357. Google Scholar CrossRef Search ADS Hothorn T, Bretz F, Westfall P. 2008. Simultaneous inference in general parametric models. Biom J . 50: 346– 363. Google Scholar CrossRef Search ADS PubMed Hopkins AD. 1917. A discussion of C.G. Hewitt’s paper on ‘insect behavior.’ J Econ Entomol . 10: 92– 93. Janz N, Söderlind L, Nylin S. 2009. No effect of larval experience on adult host preferences in Polygonia c-album (Lepidoptera: Nymphalidae): On the persistence of Hopkins’ host selection principle. Ecol Entomol . 34: 50– 57. Google Scholar CrossRef Search ADS König K., Krimmer E, Brose S, Gantert C, Buschluter I, König C, Klopfstein S, Wendt I, Baur H, Krogmann L, Steidle JLM. 2014. Does early learning drive ecological divergence during speciation processes in parasitoid wasps? Philos Trans R Soc Lond B Biol Sci . 282: 20141850– 20141850. Google Scholar CrossRef Search ADS Kos M, Houshyani B, Wietsma R, Kabouw P, Vet LE, van Loon JJ, Dicke M. 2012. Effects of glucosinolates on a generalist and specialist leaf-chewing herbivore and an associated parasitoid. Phytochemistry . 77: 162– 170. Google Scholar CrossRef Search ADS PubMed Lhomme P, Carrasco D, Larsson M, Hansson B, Anderson P. 2017. Data from: a context-dependent induction of natal habitat preference in a generalist herbivorous insect. Dryad Digital Repository . http://dx.doi.org/10.5061/dryad.rr460. Mader BJ, Daoust SP, Cardinal-Aucoin M, Bauce E, Despland E. 2012. Larval experience induces adult aversion to rearing host plants: A novel behaviour contrary to Hopkins’ host selection principle. Environ Entomol 37: 204– 211. Mery F, Burns JG. 2010. Behavioural plasticity: An interaction between evolution and experience. Evol. Ecol . 24: 571– 583. Google Scholar CrossRef Search ADS Piper WH, Palmer MW, Banfield N, Meyer MW. 2013. Can settlement in natal-like habitat explain maladaptive habitat selection? Proc Biol Sci . 280: 20130979. Google Scholar CrossRef Search ADS PubMed Plaçais PY, Preat T. 2013. To favor survival under food shortage, the brain disables costly memory. Science . 339: 440– 442. Google Scholar CrossRef Search ADS PubMed Pogue M. 2002. A World Revision of the Genus Spodoptera Guenée (Lepidoptera: Noctuidae). Mem Am Entomol Soc. Philadelphia . 202 p Price TD, Qvarnström A, Irwin DE. 2003. The role of phenotypic plasticity in driving genetic evolution. Proc R Soc Lond B Biol Sci 270: 1433– 1440. Google Scholar CrossRef Search ADS Proffit M, Khallaf MA, Carrasco D, Larsson MC, Anderson P. 2015. “Do you remember the first time?” Host plant preference in a moth is modulated by experiences during larval feeding and adult mating. Ecol Lett . 18: 365– 374 Google Scholar CrossRef Search ADS PubMed R Development Core Team. 2015. R: A Language and Environment for Statistical Computing . Vienna (Austria): R Foundation for Statistical Computing. Ray S. 1999. Survival of olfactory memory through metamorphosis in the fly Musca domestica. Neurosci Lett . 259: 37– 40. Google Scholar CrossRef Search ADS PubMed Salloum A, Colson V, Marion-Poll F. 2011. Appetitive and aversive learning in Spodoptera littoralis larvae. Chem Senses . 36: 725– 731. Google Scholar CrossRef Search ADS PubMed Selonen V, Hanski IK, Desrochers A. 2007. Natal habitat-biased dispersal in the Siberian flying squirrel. Proc R Soc Lond B Biol Sci . 274: 2063– 2068. Google Scholar CrossRef Search ADS Shields VD, Smith KP, Arnold NS, Gordon IM, Shaw TE, Waranch D. 2008. The effect of varying alkaloid concentrations on the feeding behavior of gypsy moth larvae, Lymantria dispar (L.) (Lepidoptera: Lymantriidae). Arthropod Plant Interact . 2: 101– 107. Google Scholar CrossRef Search ADS PubMed Snell-Rood EC. 2013. An overview of the evolutionary causes and consequences of behavioural plasticity. Animal Behav . 85: 1004– 1011. Google Scholar CrossRef Search ADS Stamps JA. 2006. The silver spoon effect and habitat selection by natal dispersers. Ecol Lett . 9: 1179– 1185. Google Scholar CrossRef Search ADS PubMed Stamps JA, Davis JM. 2006. Adaptive effects of natal experience on habitat selection by dispersers. Animal Behav . 72: 1279– 1289. Google Scholar CrossRef Search ADS Stamps JA, Krishnan VV, Willits NH. 2009. How different types of natal experience affect habitat preference. Am Nat . 174: 623– 630. Google Scholar CrossRef Search ADS PubMed Thöming G, Larsson MC, Hansson BS, Anderson P. 2013. Comparison of plant preference hierarchies of male and female moths and the impact of larval rearing hosts. Ecology . 94: 1744– 1752. Google Scholar CrossRef Search ADS PubMed Tully T, Cambiazo V, Kruse L. 1994. Memory through metamorphosis in normal and mutant Drosophila. J Neurosci . 14: 68– 74. Google Scholar PubMed Webster B, Qvarfordt E, Olsson U. 2013. Different roles for innate and learnt behavioral responses to odors in insect host location. Behav Ecol . 24: 366– 372. Google Scholar CrossRef Search ADS Zakir A, Sadek MM, Bengtsson M, Hansson BS, Witzgall P, Anderson P. 2013. Herbivore-induced plant volatiles provide associational resistance against an ovipositing herbivore. J. Ecol . 101: 410– 417. Google Scholar CrossRef Search ADS Zhang PJ, Liu SS, Wang H, Zalucki MP. 2007. The influence of early adult experience and larval food restriction on responses toward nonhost plants in moths. J Chem Ecol . 33: 1528– 1541. Google Scholar CrossRef Search ADS PubMed © The Author(s) 2017. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: email@example.com
Behavioral Ecology – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera