Eat First, Fight Later: Competitive Advantage of an Invasive Cichlid over a Native Competitor for Food Resources

Abstract

Competition for resources with invasive species can negatively impact native fauna. Invasive species often exhibit higher aggressiveness and monopolize resources through behavioral interference. However, their feeding behavior also plays a central role in invasion success. We investigated how food resource availability influences agonistic interactions between the invasive cichlid Oreochromis niloticus and the native cichlid Geophagus iporangensis. Specifically, we assessed whether the invasive species interferes with the native species’ feeding behavior. Using neutral arenas, we staged competition trials under two conditions: initially without food and subsequently with food present. The native species served as the focal animal and was exposed to either a conspecific or an invasive competitor. Results showed that native species aggressiveness toward the invasive competitor was three times higher in the absence of food. Although O. niloticus was 1.6 times more aggressive than conspecifics, its presence did not affect the native species’ feeding rate, and no behavioral interference was observed. Instead, the invasive species relied primarily on scramble competition, initiating nine of ten trials and consuming seven times more food than the native cichlid. Our findings suggest that, beyond aggression, feeding behavior and resource exploitation are key mechanisms driving competitive exclusion by invasive species.

1. Introduction

Invasive species are a major threat to biodiversity and the stability of ecosystems [1]. The presence of invasive species is associated with changes in community structure [2,3], disrupting interspecific interactions among native species with similar niches [4,5]. Globalization in recent decades has facilitated the spread of species into environments beyond their natural range [6]. Once an invasive species is established, it might utilize similar resources as native species, leading to niche overlap [7,8,9] and positioning itself as a strong competitor of native species [10,11] by presenting ecological functional similarities [12]. Thus, the loss of resources and reduced fitness in native species may lead to population decline and, in extreme cases, local extinction [2,13,14,15].

Competition may manifest in two primary forms: (1) behavioral interference (direct competition; [16]), through aggressive behaviors in territorial contest, aimed at directly excluding opponents from accessing resources; and (2) exploitative competition (indirect competition; [17,18]), with individual behaviors ensuring greater resource acquisition, such as ingesting food quickly and in greater quantities [19], without directly preventing competitors from accessing resources [2,20]. Thus, individuals who obtain more and better-quality resources have higher fitness [21,22] and can increase their reproductive success [23,24,25]. Both behavioral interference and exploitative competition can lead to competitive exclusion or habitat partitioning [13,16,26], although competitive interference affects individual fitness more intensely. The aggressiveness of invasive species plays an important role in the invasion process, often exhibiting highly aggressive behavior that facilitates resource monopolization and increases fitness concerning native species [27].

Agonistic contests during competition demand a high metabolic cost [28], and the decision and intensity of fighting with a similar-sized competitor may be influenced by the resource value [29,30,31,32]. For example, during food resource competition, the native Mexican mojarra Cichlasoma istlanum may exhibit higher levels of aggressiveness than the invasive species convict Cichlid Amatitlania nigrofasciata [33]; however, during nest defense, the convict Cichlid exhibits higher aggressiveness than the Mexican mojarra [34]. The value that the individual attributes to the resource in dispute may vary depending on the type of competitor [26,32] and the status of the competitor [35,36,37], reflecting their motivation to engage in fights [38]. More intense agonistic contests may happen with competitors with greater niche overlap (e.g., conspecifics) [39] and with competitors that impose a great apparent threat [40,41]. Nevertheless, native species may not recognize invasive species as potential competitors because they have not shared an evolutionary context, suffering lower levels of aggression in comparison to sympatric competitors [42]. As a result, aggressive invasive species may directly have advantages in obtaining limited resources over native species [43,44]. Consequently, native species may be at a competitive disadvantage when facing invasive species in resource acquisition.

In aggressive fish species, agonistic behavior is commonly associated with resource monopolization [45,46], and with individuals engaging in fights over territory [47], food [48], or reproductive resources [49]. For juvenile fish, territoriality is more often linked to feeding sites and access to better quality food resources than to reproductive success [48,50]. Consequently, competition between juvenile native and invasive species for food resources can act as a strong population regulation mechanism by indirectly affecting native fish species’ growth and the densities of prey communities [51]. Freshwater invasive species also tend to exhibit higher consumption rates than native counterparts [52], reinforcing the need to better understand the role of feeding behavior in aggressive invasive species as a key competition mechanism underlying the success of biological invasions

Thus, to evaluate the effect of food resource availability on agonistic interactions between aggressive invasive and native species, assessing the feeding behavior of each species and whether the invasive species exert behavioral interference, we utilize as invasive species model the Nile tilapia (Oreochromis niloticus) and as the native species model the pearl cichlid (Geophagus iporangensis; previously described as Geophagus brasiliensis [53]). The Nile tilapia is a highly invasive species present in several countries around the world due to its widespread use in fish farming [54,55]. Their common use in aquaculture is due to their generalist diet [56], high resistance to abiotic variations [57], reproduction with oral incubation [58], and accelerated growth [58,59]. These characteristics, along with its aggressive and territorial behavior, make the Nile tilapia a strong competitor, with the ability to competitively exclude native species [8,60]. The population of pearl cichlid has declined due to coexistence with Nile tilapia near urban centers, where habitat structural complexity is lower [61]. By sharing food resources and habitats, and being a territorial species [62], these invasive and native species may compete aggressively. The Nile tilapia exhibits higher aggressiveness and dominance over the pearl cichlid during territorial contests [63], demonstrating as good animal models to address competitive mechanisms underlying the food resource acquisition of aggressive invasive species.

We hypothesized that food availability would intensify the invasive species’ aggressiveness, leading to resource monopolization and a consequent reduction in the food intake by the native species. This study provides novel insights into the behavioral mechanism underlaying direct and indirect competition for food resources between aggressive invasive and native fish species, highlighting the importance of feeding behavior for invasion success even in highly aggressive invasive species.

2. Materials and Methods

2.1. Fish Acquisition and Holding Conditions

Pearl cichlid and Nile tilapia individuals were acquired from Polettini Fish Farm in the municipality of Mogi Mirim, São Paulo State (Brazil). We used juvenile, sexually immature individuals, as they seem to be more relevant to the process of competitive exclusion between the two species [61,63]. Species were acclimatized in monospecific stock systems (circular 300-L tanks), avoiding physical, visual, or chemical contact between species until the start of the experiment. The individuals were acclimatized for 30 days before the start of the experiments at a temperature of 25 ± 1 °C, with constant aeration provided by air pumps (100% saturation), physical and biological filtration, and a 12/12 photoperiod. Ammonia (<0.5 ppm) and nitrite (<0.5 ppm) levels were kept stable through weekly partial water changes. During this period, the individuals were fed twice a day until apparent satiety with commercial fish feed (32% crude protein; Guabi Pirá).

2.2. Experimental Design

To evaluate the effects of food resource availability on agonistic interactions between aggressive invasive and native species, and to evaluate whether invasive species induce interference competition in native species feeding behavior, we simulated a competition scenario initially without food resources, followed by their introduction (see Figure 1). We exposed the native species pearl cichlid (used as a focal animal) to an invasive competitor treatment (Nile tilapia; n = 10) or conspecific competitor treatment (pearl cichlid; n = 10).

Our experiment involved exposing a focal individual to a competitor (invasive or conspecific) in a neutral arena (experimental aquarium with no previous residence), allowing them to interact first in the absence of food (baseline stage) followed by the introduction of a food resource into the experimental aquarium (feeding stage; see Figure 1). We quantified aggressive behavior during both the basal and feeding stage, while the feeding behavior was assessed only during the feeding stage for both the focal individual and its competitors throughout the experimental trials.

2.3. Procedures

Before the experiments, both focal individuals and competitors (either invasive or conspecifics) were subjected to a 7-day individual acclimatization period to minimize the effects of previous social interaction [64]. One focal individual and one competitor were selected from the stocks, measured (standard length; cm), weighed (grams; g), and placed in separate 20-L acclimatization aquariums (40 × 20 × 10 cm). To avoid size asymmetry effect on aggressiveness during fights [65,66], we paired focal animals and competitors of similar sizes (±0.2 cm) [conspecific treatment: average size 6.40 ± 0.65 cm (paired Wilcoxon test: p = 0.401) and average mass 9.24 ± 2.97 g (paired Wilcoxon test: p = 0.5452; invasive treatment: average size 6.42 ± 0.51 cm (paired t-test: p = 0.912) and average mass 9.32 ± 2.19 g (paired t-test: p = 0.848)]. The acclimatization aquarium was maintained under conditions similar to those of the stock tanks, with food provided once daily, followed by a 24 h fasting period before the start of the experiment. On the seventh day, the focal animal and its respective competitor were removed from the acclimatization aquarium, marked (see Supplementary Materials with non-toxic ink for individual differentiation (red for the focal animal and black for the competitor), and simultaneously introduced into a neutral arena identical in condition and size to the acclimatization aquariums, but without prior residents. The baseline stage then began, allowing both individuals to interact in the absence of food resources for 15 min. After this period, food was introduced into the neutral arena (20 pellets) with an additional 20 pellets provided if all initial food was consumed. This marked the beginning of the feeding stage, in which individuals competed for food resources (see Figure 1). The videos were recorded using a GoPro Hero 6 mounted on a tripod positioned 1.5 m from the experimental aquarium.

2.4. Behavioral Analysis

We quantified the aggressive and feeding behaviors of the focal animal and its competitor of each treatment by analyzing the video recordings experiments using previously published ethograms of Nile tilapia [67] and pearl cichlid [62]. As variables of aggressive behavior, we used total aggression to assess fight intensity [66,68] and aggression latency to indicate fight motivation [69], measuring both across the baseline and feeding stages. Total aggression was calculated as the sum of four commonly exhibited aggressive behaviors: bites (targeting the head, tail, fin, median, or ventral areas of the opponent); lateral displays, involving a lateral body inclination to intimidate the opponent occurring with individuals oriented in the same or opposite directions; mouth wrestling, characterized by both fish interlocking their jaws and pushing against each other; and charge, defined as a rapid approach without physical contact. Aggression latency was defined as the time (in seconds) an individual took to exhibit the first aggressive behavior involving physical contact (bites, lateral displays, or frontal fighting). As feeding behavior variables, we used feeding latency, defined as the time (in seconds) required for an individual to ingest the first food pellet during the feeding stage [70] and ingestion rate, quantified as the ratio between the amount of food ingested and the individual’s weight (in grams) (adapted from [71]), allowing for comparisons of food intake across individuals of different weights.

2.5. Statistical Analysis

The data were tested for normality and homoscedasticity, with a Shapiro–Wilk and Levene test. Since the data were considered non-parametric and exhibited dependency due to the stages of the experiment or two individuals in the same aquarium, we used generalized estimating equations (GEE) to analyze both aggressive and feeding behavior response variables. This analysis consists of a marginal model for correlated structure with non-normally distributed data, being a suitable approach for behavior analysis with repeated measures and dependent data (for further detail, see [72]). GEE analysis was performed using the ‘geepack’ package [73] in R version 4.4.0 [74].

To test the effect of food resource availability on agonistic interaction (response variables: total aggression and aggression latency), we analyzed focal animal and competitor aggressive behavior separately, using as fixed factors the experimental stages (2 levels: basal and feeding stages) and type of competitor (treatments; 2 levels: conspecific and invasive species). By that, we excluded dependent data by analyzing aggressive variables from the focal animal to competitors and from competitors to the focal animal, but still had repeated measures due to the experiment stage. We selected regression models with exchangeable structures with the Poisson family and Logit function using the experiment stage as repeated measures (‘id’ function). For feeding behavior (response variables: feeding latency and ingestion rate), we used as fixed factors the type of competitor (treatment; 2 levels: conspecific and invasive species) and the individual observed (2 levels: focal animal and competitor). Thus, we have dependent data but no repeated measures. We selected regression models with exchangeable structures with the Poisson family and the Logit function using the observed individuals as dependent data (id’ function). The exchangeable correlation structure (“Ar1” function) was used for all analyses due to repeated measures (aggressive behavior variables) or dependent data (feeding behavior variables) on the same individual [72]. Also, we used the Wald test to test the significance of fixed factors, and the post hoc Tukey test was applied to assess differences between factor levels when observing differences in factor interaction. The statistical significance was considered when p < 0.05.

3. Results

3.1. Aggressive Behaviors

The focal animal exhibited similar aggression latency regardless of competitor type (GEE; Wald test, χ² = 1.45, df = 1, p = 0.229) and experimental stage (GEE; Wald test, χ² = 0.00, df = 1, p = 0.989). However, competitors displayed significantly shorter aggression latency during the baseline stage (GEE; Wald test, χ² = 19.28, df = 1, p < 0.001; Figure 2a), initiating confrontations 2.5 times faster in the absence of food resources (baseline stage: mean = 211.3 s, standard error = 38.0 s; feeding stage: mean = 539.4 s, standard error = 79.3 s). Despite this difference between experimental stages, aggression latency did not vary between competitor types (GEE; Wald test, χ² = 0.11, df = 1, p = 0.74; Figure 2a).

Despite the time to start fighting, the total aggression exhibited by the focal animal showed a significant interaction between the competitor type and the experimental stage (GEE; Wald test, χ² = 18.39, df = 1, p < 0.001). Specifically, aggression toward the invasive competitor was three times higher during the baseline stage (mean = 60.4; standard error = 7.78) compared to the feeding stage (mean = 20.3; standard error = 3.84) (Tukey: Z = 6.40; p < 0.001; Figure 2b). In the presence of food, the focal animal exhibited three times less aggression toward the invasive species (mean = 20.3; standard error = 3.84) than toward its conspecific competitor (mean = 65.5; standard error = 16.7) (Tukey: Z = −3.54; p = 0.002; Figure 2b). In contrast, the competitor’s aggression toward the focal animal did not show a significant interaction between the competitor type and experimental stage (GEE; Wald test, χ² = 0.14, df = 1, p = 0.712). However, the invasive competitor (mean = 41.7; standard error = 6.33) was 1.6 times more aggressive than the conspecific competitor (mean = 24.9; standard error = 4.97) across all experimental stages (GEE; Wald test, χ² = 3.89, df = 1, p = 0.049; Figure 2c). Additionally, overall aggression levels were twice as high during the basal stage (mean = 46; standard error = 5.87) compared to the feeding stage (mean = 20.1; standard error = 4.49) (GEE; Wald test, χ² = 18.21, df = 1, p < 0.001; Figure 2d).

3.2. Feeding Behaviors

When food resources were available in the experimental aquarium, individuals in the invasive competitor treatment fed five times faster (mean = 57.1 s; standard error = 13.6 s) compared to those in the conspecific treatment (mean = 314.1; standard error = 76.9 s) (GEE; Wald test, χ² = 18.15, df = 1, p < 0.001; Figure 3a). No significant difference was observed between the focal animal and its respective competitor in each contest (GEE; Wald test, χ² = 0.04, df = 1, p = 0.83).

Food intake exhibited a significant interaction between the type of competitor and the observed individual (GEE; competitor: stage; Wald test, χ² = 11.0, df = 1, p < 0.001; Figure 3b). The invasive competitor of the invasive treatment (mean = 2.58 pellets ingested/g; standard error = 0.345 pellets ingested/g) consumed a remarkable seven times more food than the focal animal of the invasive treatment (mean = 0.361 pellets ingested/g; standard error = 0.060 pellets ingested/g) (Tukey: Z = 13.81, p < 0.001; Figure 3b), six times more than the focal animal of the conspecific treatment (mean = 0.433 pellets ingested/g; standard error = 0.126 pellets ingested/g) (Tukey: Z = 5.88, p < 0.001; Figure 3b), and five times more than the conspecific competitor of the conspecific treatment (mean = 0.518 pellets ingested/g; standard error = 0.189 pellets ingested/g) (Tukey: Z = 4.38, p < 0.001; Figure 3b). However, no significant differences were observed between the focal animal of the invasive treatment, the focal animal of the conspecific treatment, and the conspecific competitor of the conspecific treatment.

4. Discussion

Our study demonstrated that the presence of food resources reduced the intensity of agonistic interactions with the invasive competitor and did not influence the amount of food consumed by the native species. Thus, behavioral interference was absent during food competition with the invasive competitor in our experiment. Additionally, we observed a faster feeding initiation for both focal animals and the competitors in the invasive competitor treatment, without affecting the food intake of the native species, although the invasive competitor ingested a considerably larger quantity of food. Thus, our results demonstrated behavioral flexibility in response to the presence of food resources, minimizing energy costs and optimizing resource acquisition. Our findings suggest that feeding behavior plays a key role in the competitive exclusion process even for aggressive species, indicating that exploitative competition may contribute directly to differences in native and invasive species population growth.

The increased aggressiveness observed in the absence of food is possibly related to an attempt to establish a territory, given that we used a neutral arena with the same dimensions as the acclimatization tanks. Despite competitors presenting a quicker initiation of fights during the baseline stage, the focal animals used as native species started four of ten trials against the invasive competitors, indicating similar motivation to engage in fights (Chi-square test; χ² = 0.4, df = 1, p = 0.5271). Although cichlid fish exhibit territoriality primarily associated with reproductive behaviors and intense aggression to monopolize such areas [75], territoriality in juveniles is primarily linked to feeding sites, where aggressive interactions serve to maximize food acquisition [50]. Since the experimental aquarium was a neutral arena (without prior residence), it is plausible that the individuals attempted to establish a hierarchy to ensure better access to food resources during the baseline phase of our experiment. Competition for limited resources, such as the territory of the neutral arena, is often associated with heightened aggressive behaviors as a means of monopolizing the resources [48,76]. Despite the increased aggressiveness observed during the baseline stage of our experiment, the feeding behavior of the native species was not directly affected by the presence of the invasive species, exhibiting similar ingestion rates independently of the competitor present. Therefore, our initial hypothesis of interference competition driven by the aggressiveness of the invasive species was not supported, suggesting a higher influence of resource exploitation during food resource acquisition. However, the reduction of aggressive behaviors during the feeding stage against the invasive competitor may have allowed the invasive species to obtain significantly more food than the native species. Similar findings have been reported in competition for food resources between the native European mudminnow (Umbra krameri) and the invasive Amur sleeper (Perccottus glenii), where the native fish exhibited comparable feeding rates despite taking longer to initiate feeding [77].

In our study, the invasive species appeared to exhibit behavioral flexibility in response to the availability of food resources, shifting the focus of aggressive interactions from baseline stage aggression to resource acquisition in the feeding stage. Although behavioral interference did not occur in our experiment, the feeding behavior of the invasive species highlights the importance of exploitative competition as a mechanism contributing to competitive exclusion. Exploitative competition does not have a direct effect on native species’ fitness in the same way as behavioral interference but influences population dynamics by reducing the quantity and quality of available resources for native fauna [2,16,78]. Impressively, the invasive species ingested sevenfold more food resources than the native species in the contests (focal animal), suggesting that the Nile tilapia may better exploit and use food resources than the native species. Our findings support the Resource Consumption Hypothesis, which suggests higher consumption rates by invasive species than native species of similar trophic niches [52].

Individual characteristics of the invasive species, such as higher food intake and greater energy efficiency compared to native competitors, may facilitate better growth and, consequently, enhanced reproductive performance [79,80], leading to the invasive species outpacing the native species in terms of proliferation. For example, Damas-Moreira et al. [19] observed exploitative competition between the invasive lizard Podarcis siculus and the native lizard Podarcis virescens, where the invasive competitor reached the food source more rapidly, consumed more food, and consequently gained more weight than the native species, partially explaining the displacement of the native species from its habitat. Our results suggest this may also happen to the Nile tilapia and pearl cichlid competition for food resources. Thus, in a natural competition situation, with limited resources, the Nile tilapia may be able to better obtain food resources than the pearl cichlid. In natural feeding areas, the increased consumption by invasive species may reduce the availability of food resources for native species, potentially compromising their fitness and leading to a decline in native populations [81,82]. We highlighted the significant role of feeding behavior in the competition between invasive and native aggressive species, emphasizing the importance of feeding behavior in the success of biological invasions.

The reduction of the aggressiveness by the invasive species appears to have shifted to a scramble competition strategy as a way of limiting food access to the native species without the need to use high levels of energy expenditure through aggression [83,84]. Scramble competition does not monopolize the resources but affects the opponent’s fitness by obtaining more food, more quickly, and in larger quantities [85], as previously reported for Nile tilapia [86]. The use of scramble competition instead of food resource monopolization through aggressive behavior appears to be associated with the unpredictable availability of food items [48], as observed in our experiment. In turn, feeding motivation in the invasive treatment was higher than in the conspecific treatment, as indicated by the lower feeding latency. Despite this difference between treatments, the focal animals and competitors within each treatment exhibited similar feeding motivation during the trials. Nevertheless, in the invasive treatment, the invasive competitor began feeding in 9 out of 10 trials (Chi-square test; χ² = 6.4, df = 1, p = 0.0114), suggesting that the beginning of feeding by one of the individuals may influence the opponent’s foraging efforts during competition due to social facilitation [87,88]. Thus, the native species secured the acquisition of resources during competition for food, but in smaller quantities compared to the invasive competitor.

5. Conclusions

In summary, our results show that behavioral interference did not occur during feeding competition, as the aggressiveness of the invasive species did not affect the feeding rate of the native species. In the presence of food, the invasive species has reduced its aggressiveness, engaging in scramble competition as the primary strategy to maximize food intake, with significantly higher ingestion rates than the native species. Our findings highlight that feeding behavior, resource exploitation, and exploitative competition play a crucial role in the competitive exclusion of native fauna, as they exhibit a superior ability to acquire food resources compared to native species with a similar trophic niche.