Shared shelter preferences in an endemic and an invasive catfish (Ictalurus balsanus and Ictalurus punctatus)

Shared shelter preferences in an endemic and an invasive catfish (Ictalurus balsanus and Ictalurus punctatus) | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Shared shelter preferences in an endemic and an invasive catfish (Ictalurus balsanus and Ictalurus punctatus) Josué Heredia-Mejía, Rosa Silva-Aguilar, Elsah Arce, Guillermina Alcaraz, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9124421/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Shelters are an important component of the spatial niche of many freshwater fishes, providing protection from predators and environmental stressors and influencing patterns of habitat use. When native and invasive species rely on similar refuges, overlap in shelter use may promote competition and affect the structure of aquatic communities. Here, we compared the shelter preferences of juvenile individuals of the endemic Balsas catfish Ictalurus balsanus and the invasive channel catfish I. punctatus . Each individual was presented with four shelter types (Rock, Wood, Algae, and PVC) in a sequential rank-choice experiment that generated a preference hierarchy. Both species exhibited a consistent preference structure. Wood was most frequently selected as the first refuge, followed by Rock, whereas Algae and especially PVC were rarely chosen. Analyses based on full ranking data likewise revealed a strong preference hierarchy (Wood > Rock > Algae > PVC). Despite the invasive status of I. punctatus , shelter preference patterns were broadly similar between species. These results indicate substantial overlap in refuge use between the endemic and invasive catfishes. Because structural refuges can influence survival, resting behavior, and exposure to predators, such overlap could promote competition when preferred shelters are limited in natural habitats. Understanding how native and invasive species use refuges may therefore help clarify potential mechanisms of interaction and inform management strategies in invaded freshwater ecosystems. biological invasions habitat selection shelter use conditional logistic regression Figures Figure 1 Figure 2 Introduction Interspecific competition is a key process shaping the structure of biological communities within ecosystems (Smith and Smith 2007). When species share ecological requirements, competition for limited resources can generate energetic costs that reduce feeding efficiency, growth, and reproductive success, ultimately affecting fitness. To reduce the costs associated with competition, species may exploit different portions of the ecological niche, a process known as niche partitioning. Such partitioning can occur along several niche axes, including spatial, trophic, and temporal dimensions, allowing species with similar ecological traits to coexist within the same environment (Schoener 1974 ; Vacher et al. 2016 ). In freshwater ecosystems, structural refuges are an important component of the spatial niche. Shelters can reduce exposure to predators, provide resting sites, and buffer individuals from environmental stressors such as water flow (Berryman and Hawkins 2006 ; Valdimarsson and Metcalfe 1998 ; Millidine et al. 2006 ). As a result, the availability and characteristics of refuges can strongly influence habitat use and local distribution in many benthic fishes. When structural refuges are limited, competition for shelter may become an important mechanism structuring interactions among species that occupy similar habitats. In such cases, differences in refuge selection may facilitate coexistence through spatial niche partitioning, whereas strong overlap in refuge use may increase the likelihood of competitive interactions (Van Kessel et al. 2011 ; Champneys et al. 2021 ). The study of competition between native and introduced species has received considerable attention in invasion ecology. Freshwater ecosystems are particularly susceptible to biological invasions and currently harbor a large number of introduced species worldwide (Cohen 2002 ). When invasive species share ecological traits with native species, competition for space, food, or refuges may occur, potentially altering habitat use patterns or behavioral strategies. In some cases, native species may respond to the presence of invaders by modifying their behavior, including changes in activity patterns or habitat selection, in order to reduce the frequency of encounters with competitors (Kronfeld-Schor and Dayan 2003 ; Harwood et al. 2001; Blanchet et al. 2008 ; Cucherousset and Olden 2011 ). In rivers of the Balsas basin in central Mexico, the endemic Balsas catfish Ictalurus balsanus coexists with the channel catfish Ictalurus punctatus , a species native to North America that has been widely introduced through aquaculture and fisheries activities and is now considered invasive in several regions of Mexico (Mejía-Mojica et al. 2013 ; Lara-Rivera et al. 2015 ). Both species are benthic ictalurids that frequently occupy structurally complex habitats and use shelters such as rocks, woody debris, or submerged vegetation. Their ecological similarities suggest a potential for overlap in resource use and therefore for interspecific competition. Previous work has shown that the presence of juvenile I. punctatus can alter the diel activity patterns of juvenile I. balsanus , reducing the contrast between nocturnal and diurnal activity (Rubio-Plascencia et al. 2025 ). Such behavioral changes suggest that interactions between the two species may influence how individuals use time and space within their habitat. Despite this potential for interaction, little is known about whether these species differ in their preferences for structural refuges. If both species preferentially select the same types of shelters, competition for refuges could arise when shelter availability is limited. Conversely, differences in refuge preference could facilitate spatial niche partitioning and reduce direct competition. Understanding the extent to which native and invasive species overlap in their use of refuges is therefore important for evaluating potential mechanisms of interaction in invaded freshwater systems. In this study we experimentally compared shelter preferences of juvenile Ictalurus balsanus and I. punctatus using a sequential rank-choice design that allowed individuals to generate complete preference hierarchies among four shelter types: Rock, Wood, Algae, and PVC. This approach informs not only the initial shelter selected by each fish but also the broader structure of preferences across multiple options (Arce and Alcaraz 2012 ). We tested whether the two species exhibit the same hierarchy of shelter preference and whether the relative strength of shelter selection differs between them. Based on the greater structural complexity and concealment provided by natural substrates, we predicted that both species would preferentially select Wood and Rock shelters over Algae and PVC. However, if the invasive species differs in microhabitat use from the endemic species, these differences should emerge as species-specific variation in shelter selection probabilities. Materials and methods Obtaining and maintaining organisms Juveniles of the Balsas catfish Ictalurus balsanus (N = 18) and the channel catfish Ictalurus punctatus (N = 18), of undifferentiated sex, were used in this study. Individuals of both species were obtained from the Unidad Piscícola “El Rosario” (Tehuixtla, Morelos, Mexico) and transported in oxygenated dark plastic bags to the Laboratorio de Acuicultura e Hidrobiología, Centro de Investigaciones Biológicas, Universidad Autónoma del Estado de Morelos (UAEM). Upon arrival, fish were acclimated for one week in fiberglass holding tanks (163 × 165 × 34 cm; ~590 L) under constant aeration (5.8 mg L⁻¹ dissolved oxygen), at 27 ± 1°C, pH ~ 7, and a 12 h light : 12 h dark photoperiod. Tanks were partially covered with shade mesh to reduce disturbance. Fish were fed ad libitum twice daily with live food (mosquito larvae and cladocerans). Uneaten food and feces were removed daily by siphoning, and partial water changes were performed as needed. Clay pots and tile shelters were provided in holding tanks to reduce aggressive interactions among individuals. Prior to behavioral trials, the total length (TL, cm) and body mass (W, g) of each fish were measured once. Shelter construction Four shelter types were constructed based on common refuge structures observed in rivers of the Balsas basin: Rock, Wood, Algae, and PVC. Rock and Wood shelters consisted of tunnel-like galleries constructed using natural rocks or woody branches, respectively. The Algae shelter consisted of dense artificial algal turf arranged to simulate submerged vegetation. The PVC shelter consisted of commercial smooth polyvinyl chloride tube. All shelters were tunnel-shaped with one entrance and one exit. The internal length of each shelter was adjusted to be approximately 10–20% longer than the total length of the focal fish to allow full body concealment while maintaining similar refuge dimensions across treatments. Refuge internal volume and available substrate space were standardized across shelter types. Shelters were mounted on 40 × 40 cm acrylic plates covered with a 1 cm layer of sterilized gravel. To prevent odor absorption and chemical cue bias, all materials were coated with acrylic spray enamel prior to assembly. After construction, shelters were submerged in 1% chlorinated water for 3 h, rinsed thoroughly, and aerated in clean water for 24 h to eliminate residual chlorine and odors. Shelter preference experiment Shelter preference was evaluated using a sequential multiple-choice design (Arce and Alcaraz 2012 ). Trials were conducted individually in circular fiberglass arenas (163 × 165 × 34 cm; ~590 L) under the same environmental conditions used during the acclimation period. The four shelter types were placed equidistant from the center of the arena, and their spatial position was randomized for each trial. Aeration lines were centrally positioned to maintain dissolved oxygen at approximately 5.8 mg L⁻¹ and temperature at 27 ± 1°C throughout the arena. The tanks were covered with mesh to prevent fish from jumping out. At the beginning of each trial, a focal individual was released at the center of the arena and allowed to explore freely. Individuals remained in the arena for 24 h with access to all four shelters. Food (live mosquito larvae and cladocerans) was provided twice daily at the center of the tank to avoid directional bias toward any particular refuge. After the first 24 h, the shelter with the greatest cumulative occupancy time was assigned rank 1 (most preferred) and removed from the arena. The individual was then allowed to choose among the remaining three shelters for an additional 24 h. The shelter with the highest occupancy during this phase was assigned rank 2 and removed. This procedure was repeated once more to obtain rank 3, while the remaining unchosen shelter was assigned rank 4. Thus, each individual generated a complete hierarchical ranking of the four shelter types. The total experimental duration for each fish was 96 h. Behavioral quantification Behavior was recorded continuously using QSee HD cameras (QTH41; 1280 × 720 resolution, 30 fps) mounted 1.8 m above the arena, allowing simultaneous observation of all shelters. Video analysis was conducted manually. For each 24 h phase, the first 15 min of every hour were analyzed, excluding the first 15 min of the initial day, which were considered an acclimation period. Shelter use was recorded when more than half of the fish’s body was inside a shelter. For each phase, the shelter with the greatest cumulative occupancy time (based on scan sampling) was considered the selected refuge for that stage of the ranking. Statistical analyses All analyses were performed in R (version 4.4.1; R Core Team, 2024 ) using RStudio (version 2024.04.2; Posit Software, 2024) using the packages PlackettLuce (Turner et al. 2020 ) and survival . Differences in body size between species were evaluated using Welch’s two-sample t-tests for total length (TL) and body mass (W). We calculated the proportion of individuals ranking each shelter type as their first choice (rank = 1). Differences in first-choice distributions between species were evaluated using a Pearson chi-square test. To evaluate the overall hierarchy of shelter preferences, we fitted a Plackett–Luce model to the complete ranking data pooled across species. This model estimates a “worth” parameter for each shelter type representing its relative probability of being selected across all ranking stages (Plackett 1975 ; Luce 1959; Turner et al. 2020 ). To analyze shelter selection probabilities while accounting for the sequential choice structure, each ranking was converted into a set of staged choice events: Stage 1 (choice among four shelters), Stage 2 (choice among three remaining shelters), and Stage 3 (choice among two shelters). We fitted conditional logistic regression models with choice-set strata and clustering by individual to account for repeated observations within fish. Two models were considered: m0: chosen ~ shelter + strata(choice_set) + cluster(individual) and m1: chosen ~ shelter × species + strata(choice_set) + cluster(individual). Odds ratios (OR) were estimated with 95% confidence intervals. Rock was treated as the reference shelter for the baseline species ( I. balsanus ). Finally, to test whether shelter preferences differed between species, we compared models m0 and m1 using a likelihood ratio test (LRT). For this comparison, models were refitted without the clustering term to obtain comparable likelihood values. Results All 36 individuals produced complete shelter rankings (ranks 1–4), and there were no missing values in species identity, individual identity, shelter choice, total length, or body mass. Total length did not differ between species (mean = 18.92 ± 1.48 cm; Welch’s t = 0.325, df = 33.72, p = 0.747). Body mass also did not differ significantly between species (mean = 41.87 ± 8.31 g; Welch’s t = − 0.638, df = 27.30, p = 0.529). Given the absence of body size differences, body size was not included as a covariate in subsequent analyses. The proportion of individuals ranking each shelter type as their first choice is shown in Fig. 1 . First-choice frequencies did not differ significantly between species (χ² = 2.333, df = 3, p = 0.506; Table 1 ). In both species, Wood was the most frequently selected shelter as first choice (10 of 18 individuals in each species; 55.6%), followed by Rock (27.8% in I. balsanus and 38.9% in I. punctatus ). Algae was occasionally ranked first by I. balsanus (11.1%) but not by I. punctatus , whereas PVC was rarely selected as the first choice in either species (5.6% in both cases). Table 1 First-choice (rank = 1) shelter frequencies and proportions by species. Species Rock n (%) Wood n (%) Algae n (%) PVC n (%) I. balsanus (n = 18) 5 (27.8) 10 (55.6) 2 (11.1) 1 (5.6) I. punctatus (n = 18) 7 (38.9) 10 (55.6) 0 (0.0) 1 (5.6) The complete ranking data also showed a clear overall hierarchy of shelter preference. The Plackett–Luce model fitted to rankings pooled across species assigned the highest worth to Wood (0.552), followed by Rock (0.320), Algae (0.084), and PVC (0.043). Thus, Wood and Rock together accounted for most of the overall preference weight, whereas Algae and especially PVC were consistently less preferred. Conditional-choice analyses also supported a strong effect of shelter type on selection probability (Table 2 ; Fig. 2 ). For the baseline species, I. balsanus , Wood showed higher odds of being selected than Rock (OR = 2.004, 95% CI = 0.791–5.076, p = 0.143), whereas Algae and PVC showed lower odds of selection than Rock (Algae: OR = 0.486, 95% CI = 0.212–1.113, p = 0.088; PVC: OR = 0.228, 95% CI = 0.098–0.530, p Rock > Algae > PVC. Table 2 Conditional-choice model results: odds ratios (OR) with 95% confidence intervals for shelter selection, including species-specific contrasts; Rock is the reference shelter for the baseline species ( I. balsanus ) Term OR 95% CI p Wood 2.004 0.791–5.076 0.143 Algae 0.486 0.212–1.113 0.0879 PVC 0.228 0.098–0.530 0.000599 Rock × I. punctatus 4.964 0.743–33.172 0.0983 Wood × I. punctatus 3.587 0.514–25.011 0.197 Algae × I. punctatus 0.918 0.228–3.699 0.904 PVC × I. punctatus NA NA NA PVC × I. punctatus was not estimable (NA), consistent with quasi-complete separation due to extremely rare selection of PVC by I. punctatus within the staged choice sets. Allowing shelter preferences to vary between species did not significantly improve model fit compared to a model assuming a shared preference structure (likelihood-ratio test: χ² = 5.822, df = 3, p = 0.121). Thus, there was no statistically robust evidence that I. balsanus and I. punctatus differed in their overall hierarchy of shelter selection. Species-specific coefficients were generally imprecisely estimated, with wide confidence intervals (Table 2 ). In addition, the PVC × I. punctatus contrast was not estimable, consistent with quasi-complete separation caused by the extremely rare selection of PVC by I. punctatus in the staged choice sets. Discussion Juvenile Ictalurus balsanus and I. punctatus exhibited a clear and consistent hierarchy of shelter preference, with Wood and Rock being selected more often than Algae and PVC. This pattern was evident across all analytical approaches used in this study, including first-choice frequencies, the Plackett–Luce ranking model, and the conditional-choice analysis. Preference tests are widely used in behavioral ecology to infer how animals value alternative options, and combining rank-based inference with staged choice models provides a particularly strong basis for concluding that selection was structured rather than random (Kirkden and Pajor 2006 ). Shelter preference is expected to have ecological consequences because refuges can mediate key processes that determine survival and local distribution. The refuge concept links sheltering behavior to predator avoidance, resource limitation, and the outcomes of species interactions (Berryman and Hawkins 2006 ). In fishes and other aquatic animals, sheltering can reduce exposure to predators and environmental stress, and shelter use can also be shaped by external conditions (e.g., seasonal constraints or light-related risk), making refuge selection a plausible axis of the spatial niche (Valdimarsson and Metcalfe 1998 ; Kerry and Bellwood 2017 ). Under this framework, the strong skew toward Wood and Rock suggests that these shelter types likely provide functional benefits that the fish can detect behaviorally, even though the adaptive value of specific shelters often requires direct tests that link choice to performance (Arce and Alcaraz 2012 ). In this study, invasive I. punctatus did not show a shift in shelter preference relative to the endemic I. balsanus . Both species exhibited broadly similar preference hierarchies, and allowing shelter preferences to vary between species did not significantly improve model fit. This implies substantial microhabitat overlap in refuge use where the two species coexist, and it reinforces concerns about competitive interactions in invaded systems given that I. punctatus is established in tributaries of the Balsas basin and is considered a conservation concern for native fishes (Mejía-Mojica et al. 2013 ). Although, “shared preference” does not necessarily mean “no impact”, in other invaded systems, competition for access to preferred shelters can occur even when species share similar habitat requirements, and invasive species can restrict native access to shelters via interference or displacement (Van Kessel et al. 2011 ; Champneys et al. 2021 ). This inference is consistent with prior evidence that juvenile I. punctatus can alter the diel activity pattern of juvenile I. balsanus , indicating that behavioral interactions between the two species can be strong even in controlled conditions (Rubio-Plascencia et al. 2025 ). Across aquatic taxa, strong shelter preferences are commonly linked to specific physical properties of refuges, which provides a mechanistic context for our Wood/Rock versus PVC pattern. For example, spiny lobsters can strongly prefer shelters that provide shade or cover and appropriate entrance configurations, and their den choice can vary with predation risk and social conditions (Spanier and Zimmer-Faust 1988 ; Eggleston and Lipcius 1992). Similarly, slipper lobsters prefer shaded or opaque shelters over transparent ones, consistent with concealment-based mechanisms (Spanier and Almog-Shtayer 1992 ). In crabs, refuge geometry can influence both preference and agonistic interactions; swimming crabs show significant shelter-shape preferences, and some configurations can reduce willingness to fight (Zhang et al. 2021 ). These cross-taxa examples support the idea that animals evaluate shelters based on how refuge design alters concealment and escape options. PVC was particularly rarely selected in our trials, a result that aligns with multiple studies showing that smooth artificial shelters may be less preferred than rougher or more complex alternatives. In Arctic charr aquaculture, a PVC-based shelter design with grooves altered behavior and performance, highlighting that fine-scale structural features of artificial refuges can matter (Benhaïm et al. 2009 ). In the zebra pleco, shelter selection in rearing conditions shows that fish can use PVC, but may show higher utilization of alternative, rougher shelters (like brick or clay) relative to smooth PVC pipes (Ramos et al. 2013 ). Outside fishes, sea cucumbers also show shelter preferences depending on material and structural design (e.g., PVC versus cement-type materials and opening characteristics), emphasizing that material effects are widespread across aquatic animals (Dong et al. 2010 ; Kwon et al. 2019 ). It is possible that surface texture, internal complexity, and perceived concealment contribute to why Wood and Rock were preferred over PVC in our catfishes. In addition to geometry and texture, visual and camouflage-related cues provide another plausible explanation for the avoidance of PVC in particular. Camouflage theory predicts that matching background color and reducing conspicuousness can reduce detection risk, and several studies show that animals can exhibit color-associated preferences in sheltering contexts (Stevens and Merilaita 2008; Pessani and Tirelli 2006 ). For example, mud crabs show shelter color preferences in controlled tests (Kawamura et al. 2020), and hermit crab shell choice can be related to chromatic resemblance between crab and occupied shell (Pessani and Tirelli 2006 ). In this study, it is plausible that the darker, heterogeneous appearance of Wood (and, depending on construction, Rock) may offer better concealment than relatively uniform, light-colored PVC, especially under daylight conditions. Shelter preference can also be influenced by trophic or chemical pathways, because some refuges accumulate biofilms or concentrate prey resources. In rivers, in-stream wood can function not only as physical cover but also as a substrate linked to food resources and habitat patchiness, making wood a biologically “active” refuge type (Pettit et al. 2013 ). In aquaculture contexts, sea cucumbers show differential responses to shelter materials and colors, and shelter design is often evaluated precisely because it changes local conditions experienced by the organism (Dong et al. 2010 ; Kwon et al. 2019 ). In our study, refuges were cleaned and treated to reduce odor absorption and chemical cue bias, so we do not interpret our results as driven by biofilm accumulation. However, in natural settings, chemical and trophic differences among refuge types could reinforce structural preferences and further increase overlap between native and invasive catfishes on the same refuge patches. From a management perspective, many restoration programs aim to increase physical complexity by maintaining rocky substrate and large wood inputs, because structural complexity is linked to habitat quality and behavioral opportunities for aquatic organisms (Tokeshi and Arakaki 2012 ; Pettit et al. 2013 ). Our results suggest that such actions would likely benefit both catfish species rather than selectively favoring the endemic fish, since both I. balsanus and invasive I. punctatus preferred the same refuge types. In addition, PVC and other artificial shelters are widely used in aquaculture and in applied contexts to mitigate aggression or cannibalism, and there is active work on screening shelter designs for different species (Benhaïm et al. 2009 ). However, our results caution against assuming that PVC is an appropriate refuge substitute for all fishes, because preference can be highly species- and context-dependent and may fail to match key physical or sensory features of natural refuges. It is worthy to note that our trials were conducted in isolation and therefore did not include direct interspecific interactions; the realized use of shelters under interference competition may differ from preference expressed without competitors. This matters because shelter competition is a documented mechanism of invasive impacts in both fishes and crustaceans, often producing exclusion from preferred shelters or dominance by one species (Van Kessel et al. 2011 ; Champneys et al. 2021 ; Vorburger and Ribi 1999 ). Additionally, some shelter–species combinations were rare (e.g., PVC by I. punctatus ), producing quasi-complete separation and limiting precision for some interaction terms, although not altering the overall hierarchy supported by multiple methods. We also used juveniles of both species; because sheltering behavior and refuge needs can change across ontogeny in some taxa, future work should test whether the same hierarchy holds across life stages. Finally, preference may shift with predation risk and social context, as demonstrated in classic shelter-selection work on lobsters (Eggleston and Lipcius 1992), suggesting that predator-cue experiments and direct competition trials are the most direct way to test the adaptive and ecological consequences of shelter choice in this system. Statements and Declarations Acknowledgements This study was supported by Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT, IA-202424). We thank Olivia de los Santos, Ignacio Morales-Salas, Maria Pérez-Cruz and Cassandra Rubio for their technical assistance in the field. Funding This study was funded by the Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT, IA-202424). Competing Interests The authors have no relevant financial or non-financial interests to disclose. Author Contributions All authors conceived the study. HMJ and SAR conducted experiments. LMB analysed the data. LMB wrote the first draft. All authors revised the manuscript and approved the final version. Declaration of generative AI During the preparation of this work, the authors used ChatGPT (OpenAI) to assist with English language checking. After using this tool, the authors reviewed the content and take full responsibility for the final manuscript. Ethics approval The study was approved by the Commission of Ethics and Scientific Responsibility, Faculty of Sciences, UNAM (CEARC/19062023 PI_06_06_2023_Burciaga). All experimental procedures in this study were performed complying with Mexican legislation. Data and Code Availability Data and code are available at figshare (https://doi.org/10.6084/m9.figshare.31724407) References Arce E, Alcaraz G (2012) Shell preference in a hermit crab: comparison between a matrix of paired comparisons and a multiple-alternative experiment. Mar Biol 159:853–862. https://doi.org/10.1007/s00227-011-1861-x Benhaïm D, Leblanc CA, Lucas G (2009) Impact of a new artificial shelter on Arctic charr ( Salvelinus alpinus L.) behaviour and culture performance during the endogenous feeding period. Aquaculture 295:38–43. https://doi.org/10.1016/j.aquaculture.2009.06.024 Berryman AA, Hawkins BA (2006) The refuge as an integrating concept in ecology and evolution. Oikos 115:192–196. https://doi.org/10.1111/j.0030-1299.2006.15188.x Blanchet S, Loot G, Bernatchez L, Dodson JJ (2008) The effects of abiotic factors and intraspecific versus interspecific competition on the diel activity patterns of Atlantic salmon ( Salmo salar ) fry. Can J Fish Aquat Sci 65:1545–1553. https://doi.org/10.1139/F08-079 Champneys T, Genner MJ, Ioannou CC (2021) Invasive Nile tilapia dominates a threatened indigenous tilapia in competition over shelter. Hydrobiologia 848:3747–3762. https://doi.org/10.1007/s10750-020-04341-8 Cohen AN, Webb SK (eds) (2002) Biological invasions in aquatic ecosystems: impacts on restoration and potential for control. San Francisco Estuary Institute, Oakland Cucherousset J, Olden JD (2011) Ecological impacts of non-native freshwater fishes. Fisheries 36:215–230. https://doi.org/10.1080/03632415.2011.574578 Dong G, Dong S, Wang F, Tian X (2010) Effects of materials, incubation time and colors of artificial shelters on behavior of juvenile sea cucumber Apostichopus japonicus . Aquacult Eng 43:1–5. https://doi.org/10.1016/j.aquaeng.2010.01.002 Kerry JT, Bellwood DR (2017) Environmental drivers of sheltering behaviour in large reef fishes. Mar Pollut Bull 125:254–259. https://doi.org/10.1016/j.marpolbul.2017.08.036 Kirkden RD, Pajor EA (2006) Using preference, motivation and aversion tests to ask scientific questions about animals’ feelings. Appl Anim Behav Sci 100:29–47. https://doi.org/10.1016/j.applanim.2006.04.009 Kronfeld-Schor N, Dayan T (2003) Partitioning of time as an ecological resource. Annu Rev Ecol Evol Syst 34:153–181. https://doi.org/10.1146/annurev.ecolsys.34.011802.132435 Kwon I, Lee K, Kim T (2019) Shelter material and shape preferences of the sea cucumber Apostichopus japonicus . Aquaculture 508:206–213. https://doi.org/10.1016/j.aquaculture.2019.05.011 Lara-Rivera AL, Parra-Bracamonte GM, Sifuentes-Rincón AM, Gojón-Báez HH, Rodríguez-González H, Montelongo-Alfaro IO (2015) El bagre de canal ( Ictalurus punctatus Rafinesque, 1818): estado actual y problemática en México. Lat Am J Aquat Res 43:424–434. https://doi.org/10.3856/vol43-issue3-fulltext-4 Mejía-Mojica H, Paredes-Lira ME, Beltrán-López RG (2013) Primer registro y establecimiento del bagre de canal Ictalurus punctatus (Siluriformes: Ictaluridae) en un tributario del Río Balsas, México. Hidrobiológica 23:456–459 Millidine KJ, Armstrong JD, Metcalfe NB (2006) Presence of shelter reduces maintenance metabolism of juvenile salmon. Funct Ecol 20:839–845. https://doi.org/10.1111/j.1365-2435.2006.01166.x Posit Team (2024) RStudio: Integrated Development Environment for R. Posit Software, Boston Pettit NE, Warfe DM, Kennard MJ, Pusey BJ, Davies PM, Douglas MM (2013) Dynamics of in-stream wood and its importance as fish habitat in a large tropical floodplain river. River Res Appl 29:864–875 Plackett RL (1975) The analysis of permutations. J R Stat Soc Ser C Appl Stat 24:193–202. https://doi.org/10.2307/2346567 Pessani D, Tirelli T (2006) Chromatic patterns of the hermit crab Calcinus tubularis related to the occupied shell. Hydrobiologia 557:107–112. https://doi.org/10.1007/s10750-005-1313-z R Core Team (2024) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna Ramos FM, Recuero LB, Silva TVN, Fujimoto RY, Lee JT, Ferreira-Torres M (2013) Shelter selection in the Amazonian zebra pleco Hypancistrus zebra . J Appl Ichthyol 29:920–927. https://doi.org/10.1111/jai.12176 Rubio-Plascencia C, Kiere LM, Burciaga LM (2025) Juveniles of the endemic catfish Ictalurus balsanus are less nocturnal in the presence of juvenile Ictalurus punctatus . Environ Biol Fishes 108:957–968. https://doi.org/10.1007/s10641-025-01700-z Schoener TW (1974) Resource partitioning in ecological communities. Science 185:27–39. https://doi.org/10.1126/science.185.4145.27 Smith TM, Smith RL (2009) Elements of ecology, 7th edn. Pearson Benjamin Cummings, San Francisco Spanier E, Zimmer-Faust RK (1988) Some physical properties of shelter that influence den preference in spiny lobsters. J Exp Mar Biol Ecol 121:137–149. https://doi.org/10.1016/0022-0981(88)90251-1 Spanier E, Almog-Shtayer G (1992) Shelter preferences in the Mediterranean slipper lobster: effects of physical properties. J Exp Mar Biol Ecol 164:103–116. https://doi.org/10.1016/0022-0981(92)90139-2 Stevens M, Merilaita S (2009) Animal camouflage: current issues and new perspectives. Philos Trans R Soc B 364:423–427. https://doi.org/10.1098/rstb.2008.0217 Tokeshi M, Arakaki S (2012) Habitat complexity in aquatic systems: fractals and beyond. Hydrobiologia 685:27–47. https://doi.org/10.1007/s10750-011-0832-z Turner HL, van Etten J, Firth D, Kosmidis I (2020) Modelling rankings in R: the PlackettLuce package. Comput Stat 35:1027–1057. https://doi.org/10.1007/s00180-020-00959-3 Vacher C, Tamaddoni-Nezhad A, Kamenova S et al (2016) Learning ecological networks from next-generation sequencing data. In: Woodward G, Bohan DA (eds) Adv Ecol Res, vol 54. Academic Press, Oxford, pp 1–39. https://doi.org/10.1016/bs.aecr.2015.10.004 Valdimarsson SK, Metcalfe NB (1998) Shelter selection in juvenile Atlantic salmon, or why do salmon seek shelter in winter? J Fish Biol 52:42–49. https://doi.org/10.1111/j.1095-8649.1998.tb01551.x van Kessel N, Dorenbosch M, de Boer MRM, Leuven RSEW, van der Velde G (2011) Competition for shelter between four invasive gobiids and two native benthic fish species. Curr Zool 57:844–851. https://doi.org/10.1093/czoolo/57.6.844 Vorburger C, Ribi G (1999) Aggression and competition for shelter between a native and an introduced crayfish in Europe. Freshw Biol 42:111–119. https://doi.org/10.1046/j.1365-2427.1999.00465.x Zhang H, Zhu B, Yu L, Liu D, Wang F, Lu Y (2021) Selection of shelter shape by swimming crab ( Portunus trituberculatus ). Aquac Rep 21:100908. https://doi.org/10.1016/j.aqrep.2021.100908 Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 09 May, 2026 Reviewers invited by journal 02 Apr, 2026 Editor invited by journal 19 Mar, 2026 Editor assigned by journal 16 Mar, 2026 First submitted to journal 14 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9124421","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":616484459,"identity":"27d88d7f-2f6d-475c-ad8c-c9410b5f3a98","order_by":0,"name":"Josué Heredia-Mejía","email":"","orcid":"","institution":"Universidad Autonoma del Estado de Morelos","correspondingAuthor":false,"prefix":"","firstName":"Josué","middleName":"","lastName":"Heredia-Mejía","suffix":""},{"id":616484460,"identity":"f1225c06-b09c-4c9c-94ff-0bf4633d5cf8","order_by":1,"name":"Rosa Silva-Aguilar","email":"","orcid":"","institution":"Universidad Nacional Autonoma de Mexico Facultad de Ciencias","correspondingAuthor":false,"prefix":"","firstName":"Rosa","middleName":"","lastName":"Silva-Aguilar","suffix":""},{"id":616484461,"identity":"9203bc28-e11d-4036-80d7-39dec9795ab0","order_by":2,"name":"Elsah Arce","email":"","orcid":"","institution":"Universidad Autonoma del Estado de Morelos","correspondingAuthor":false,"prefix":"","firstName":"Elsah","middleName":"","lastName":"Arce","suffix":""},{"id":616484462,"identity":"f6151b5c-3a9e-453c-8702-f5ad4914b5bf","order_by":3,"name":"Guillermina Alcaraz","email":"","orcid":"","institution":"Universidad Nacional Autonoma de Mexico Facultad de Ciencias","correspondingAuthor":false,"prefix":"","firstName":"Guillermina","middleName":"","lastName":"Alcaraz","suffix":""},{"id":616484463,"identity":"451d43e6-c91a-4528-a339-3ee834571706","order_by":4,"name":"Luis M. Burciaga","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAklEQVRIiWNgGAWjYBACAxCRAMT8DIwNQEqCBC2SDSRpATMOEOswc4nchx8e1NTKGx8/3Pi4oMZCXr6Bx+zBzxyGaFyGWM5IN5ZIOHbccNuZxGbjGcckDDcc4DE37N3GkDuzAYfDbqSxMSSwHWPcdoOxTZq3QYJxAwOPmQQvUEs/Lr+Atfw7Zr95BmP7b6AW+/lAh0n+BWppw6clsa0mcYMEYxszUEtiwwEeM2l8tlj2PGOWSOw7kDwD6BdpnmMSyRsOs5VJy26TwOkXc/Y0xo8/vtXZ9rcff/iZp6bOdn578zbJt9tscjfgCDEoOIzEZgaTBOO0jpCCUTAKRsEoGMkAAMcaV54S/OhbAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0003-4094-3663","institution":"Universidad Nacional Autonoma de Mexico Facultad de Ciencias","correspondingAuthor":true,"prefix":"","firstName":"Luis","middleName":"M.","lastName":"Burciaga","suffix":""}],"badges":[],"createdAt":"2026-03-14 18:21:45","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9124421/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9124421/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106392869,"identity":"cfa03b10-d779-47a8-b551-8fbb0424d9ce","added_by":"auto","created_at":"2026-04-08 07:29:12","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":38347,"visible":true,"origin":"","legend":"\u003cp\u003eProportion of individuals ranking each shelter type as their first choice (rank = 1) for \u003cem\u003eIctalurus balsanus\u003c/em\u003e and \u003cem\u003eI. punctatus\u003c/em\u003e. Wood was most frequently selected in both species, followed by Rock, whereas Algae and PVC were rarely ranked first\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9124421/v1/3d119b2d4403aecb5d9e573e.png"},{"id":106392974,"identity":"0aaa4e3d-6f56-4cd8-9876-22555d92e45d","added_by":"auto","created_at":"2026-04-08 07:29:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":58064,"visible":true,"origin":"","legend":"\u003cp\u003eOdds ratios (log scale) from the conditional-choice model. Points show OR estimates and horizontal lines show 95% confidence intervals. Values \u0026gt; 1 indicate higher odds of selection relative to the Rock reference (baseline species \u003cem\u003eI. balsanus\u003c/em\u003e); species contrasts are shown as shelter-specific terms for \u003cem\u003eI. punctatus\u003c/em\u003e. The vertical dashed line indicates OR = 1\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9124421/v1/ea9b87d29e2a2244211f1e63.png"},{"id":106393041,"identity":"d13e97ca-c467-4c16-a6bd-80b854be2ea8","added_by":"auto","created_at":"2026-04-08 07:29:54","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":636731,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9124421/v1/4033e322-ca37-4ae3-9012-f2dd5ed6d04a.pdf"}],"financialInterests":"","formattedTitle":"Shared shelter preferences in an endemic and an invasive catfish (Ictalurus balsanus and Ictalurus punctatus)","fulltext":[{"header":"Introduction","content":"\u003cp\u003eInterspecific competition is a key process shaping the structure of biological communities within ecosystems (Smith and Smith 2007). When species share ecological requirements, competition for limited resources can generate energetic costs that reduce feeding efficiency, growth, and reproductive success, ultimately affecting fitness. To reduce the costs associated with competition, species may exploit different portions of the ecological niche, a process known as niche partitioning. Such partitioning can occur along several niche axes, including spatial, trophic, and temporal dimensions, allowing species with similar ecological traits to coexist within the same environment (Schoener \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e1974\u003c/span\u003e; Vacher et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn freshwater ecosystems, structural refuges are an important component of the spatial niche. Shelters can reduce exposure to predators, provide resting sites, and buffer individuals from environmental stressors such as water flow (Berryman and Hawkins \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Valdimarsson and Metcalfe \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e1998\u003c/span\u003e; Millidine et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). As a result, the availability and characteristics of refuges can strongly influence habitat use and local distribution in many benthic fishes. When structural refuges are limited, competition for shelter may become an important mechanism structuring interactions among species that occupy similar habitats. In such cases, differences in refuge selection may facilitate coexistence through spatial niche partitioning, whereas strong overlap in refuge use may increase the likelihood of competitive interactions (Van Kessel et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Champneys et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe study of competition between native and introduced species has received considerable attention in invasion ecology. Freshwater ecosystems are particularly susceptible to biological invasions and currently harbor a large number of introduced species worldwide (Cohen \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). When invasive species share ecological traits with native species, competition for space, food, or refuges may occur, potentially altering habitat use patterns or behavioral strategies. In some cases, native species may respond to the presence of invaders by modifying their behavior, including changes in activity patterns or habitat selection, in order to reduce the frequency of encounters with competitors (Kronfeld-Schor and Dayan \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Harwood et al. 2001; Blanchet et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Cucherousset and Olden \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn rivers of the Balsas basin in central Mexico, the endemic Balsas catfish \u003cem\u003eIctalurus balsanus\u003c/em\u003e coexists with the channel catfish \u003cem\u003eIctalurus punctatus\u003c/em\u003e, a species native to North America that has been widely introduced through aquaculture and fisheries activities and is now considered invasive in several regions of Mexico (Mej\u0026iacute;a-Mojica et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Lara-Rivera et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Both species are benthic ictalurids that frequently occupy structurally complex habitats and use shelters such as rocks, woody debris, or submerged vegetation. Their ecological similarities suggest a potential for overlap in resource use and therefore for interspecific competition. Previous work has shown that the presence of juvenile \u003cem\u003eI. punctatus\u003c/em\u003e can alter the diel activity patterns of juvenile \u003cem\u003eI. balsanus\u003c/em\u003e, reducing the contrast between nocturnal and diurnal activity (Rubio-Plascencia et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Such behavioral changes suggest that interactions between the two species may influence how individuals use time and space within their habitat.\u003c/p\u003e \u003cp\u003eDespite this potential for interaction, little is known about whether these species differ in their preferences for structural refuges. If both species preferentially select the same types of shelters, competition for refuges could arise when shelter availability is limited. Conversely, differences in refuge preference could facilitate spatial niche partitioning and reduce direct competition. Understanding the extent to which native and invasive species overlap in their use of refuges is therefore important for evaluating potential mechanisms of interaction in invaded freshwater systems.\u003c/p\u003e \u003cp\u003eIn this study we experimentally compared shelter preferences of juvenile \u003cem\u003eIctalurus balsanus\u003c/em\u003e and \u003cem\u003eI. punctatus\u003c/em\u003e using a sequential rank-choice design that allowed individuals to generate complete preference hierarchies among four shelter types: Rock, Wood, Algae, and PVC. This approach informs not only the initial shelter selected by each fish but also the broader structure of preferences across multiple options (Arce and Alcaraz \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). We tested whether the two species exhibit the same hierarchy of shelter preference and whether the relative strength of shelter selection differs between them. Based on the greater structural complexity and concealment provided by natural substrates, we predicted that both species would preferentially select Wood and Rock shelters over Algae and PVC. However, if the invasive species differs in microhabitat use from the endemic species, these differences should emerge as species-specific variation in shelter selection probabilities.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eObtaining and maintaining organisms\u003c/h2\u003e \u003cp\u003eJuveniles of the Balsas catfish \u003cem\u003eIctalurus balsanus\u003c/em\u003e (N\u0026thinsp;=\u0026thinsp;18) and the channel catfish \u003cem\u003eIctalurus punctatus\u003c/em\u003e (N\u0026thinsp;=\u0026thinsp;18), of undifferentiated sex, were used in this study. Individuals of both species were obtained from the Unidad Pisc\u0026iacute;cola \u0026ldquo;El Rosario\u0026rdquo; (Tehuixtla, Morelos, Mexico) and transported in oxygenated dark plastic bags to the Laboratorio de Acuicultura e Hidrobiolog\u0026iacute;a, Centro de Investigaciones Biol\u0026oacute;gicas, Universidad Aut\u0026oacute;noma del Estado de Morelos (UAEM).\u003c/p\u003e \u003cp\u003eUpon arrival, fish were acclimated for one week in fiberglass holding tanks (163 \u0026times; 165 \u0026times; 34 cm; ~590 L) under constant aeration (5.8 mg L⁻\u0026sup1; dissolved oxygen), at 27\u0026thinsp;\u0026plusmn;\u0026thinsp;1\u0026deg;C, pH\u0026thinsp;~\u0026thinsp;7, and a 12 h light : 12 h dark photoperiod. Tanks were partially covered with shade mesh to reduce disturbance. Fish were fed ad libitum twice daily with live food (mosquito larvae and cladocerans). Uneaten food and feces were removed daily by siphoning, and partial water changes were performed as needed. Clay pots and tile shelters were provided in holding tanks to reduce aggressive interactions among individuals. Prior to behavioral trials, the total length (TL, cm) and body mass (W, g) of each fish were measured once.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eShelter construction\u003c/h3\u003e\n\u003cp\u003eFour shelter types were constructed based on common refuge structures observed in rivers of the Balsas basin: Rock, Wood, Algae, and PVC. Rock and Wood shelters consisted of tunnel-like galleries constructed using natural rocks or woody branches, respectively. The Algae shelter consisted of dense artificial algal turf arranged to simulate submerged vegetation. The PVC shelter consisted of commercial smooth polyvinyl chloride tube.\u003c/p\u003e \u003cp\u003eAll shelters were tunnel-shaped with one entrance and one exit. The internal length of each shelter was adjusted to be approximately 10\u0026ndash;20% longer than the total length of the focal fish to allow full body concealment while maintaining similar refuge dimensions across treatments. Refuge internal volume and available substrate space were standardized across shelter types.\u003c/p\u003e \u003cp\u003eShelters were mounted on 40 \u0026times; 40 cm acrylic plates covered with a 1 cm layer of sterilized gravel. To prevent odor absorption and chemical cue bias, all materials were coated with acrylic spray enamel prior to assembly. After construction, shelters were submerged in 1% chlorinated water for 3 h, rinsed thoroughly, and aerated in clean water for 24 h to eliminate residual chlorine and odors.\u003c/p\u003e\n\u003ch3\u003eShelter preference experiment\u003c/h3\u003e\n\u003cp\u003eShelter preference was evaluated using a sequential multiple-choice design (Arce and Alcaraz \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Trials were conducted individually in circular fiberglass arenas (163 \u0026times; 165 \u0026times; 34 cm; ~590 L) under the same environmental conditions used during the acclimation period.\u003c/p\u003e \u003cp\u003eThe four shelter types were placed equidistant from the center of the arena, and their spatial position was randomized for each trial. Aeration lines were centrally positioned to maintain dissolved oxygen at approximately 5.8 mg L⁻\u0026sup1; and temperature at 27\u0026thinsp;\u0026plusmn;\u0026thinsp;1\u0026deg;C throughout the arena. The tanks were covered with mesh to prevent fish from jumping out.\u003c/p\u003e \u003cp\u003eAt the beginning of each trial, a focal individual was released at the center of the arena and allowed to explore freely. Individuals remained in the arena for 24 h with access to all four shelters. Food (live mosquito larvae and cladocerans) was provided twice daily at the center of the tank to avoid directional bias toward any particular refuge.\u003c/p\u003e \u003cp\u003eAfter the first 24 h, the shelter with the greatest cumulative occupancy time was assigned rank 1 (most preferred) and removed from the arena. The individual was then allowed to choose among the remaining three shelters for an additional 24 h. The shelter with the highest occupancy during this phase was assigned rank 2 and removed. This procedure was repeated once more to obtain rank 3, while the remaining unchosen shelter was assigned rank 4. Thus, each individual generated a complete hierarchical ranking of the four shelter types. The total experimental duration for each fish was 96 h.\u003c/p\u003e\n\u003ch3\u003eBehavioral quantification\u003c/h3\u003e\n\u003cp\u003eBehavior was recorded continuously using QSee HD cameras (QTH41; 1280 \u0026times; 720 resolution, 30 fps) mounted 1.8 m above the arena, allowing simultaneous observation of all shelters.\u003c/p\u003e \u003cp\u003eVideo analysis was conducted manually. For each 24 h phase, the first 15 min of every hour were analyzed, excluding the first 15 min of the initial day, which were considered an acclimation period. Shelter use was recorded when more than half of the fish\u0026rsquo;s body was inside a shelter. For each phase, the shelter with the greatest cumulative occupancy time (based on scan sampling) was considered the selected refuge for that stage of the ranking.\u003c/p\u003e\n\u003ch3\u003eStatistical analyses\u003c/h3\u003e\n\u003cp\u003eAll analyses were performed in \u003cem\u003eR\u003c/em\u003e (version 4.4.1; R Core Team, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) using \u003cem\u003eRStudio\u003c/em\u003e (version 2024.04.2; Posit Software, 2024) using the packages \u003cem\u003ePlackettLuce\u003c/em\u003e (Turner et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and \u003cem\u003esurvival\u003c/em\u003e. Differences in body size between species were evaluated using Welch\u0026rsquo;s two-sample t-tests for total length (TL) and body mass (W). We calculated the proportion of individuals ranking each shelter type as their first choice (rank\u0026thinsp;=\u0026thinsp;1). Differences in first-choice distributions between species were evaluated using a Pearson chi-square test. To evaluate the overall hierarchy of shelter preferences, we fitted a Plackett\u0026ndash;Luce model to the complete ranking data pooled across species. This model estimates a \u0026ldquo;worth\u0026rdquo; parameter for each shelter type representing its relative probability of being selected across all ranking stages (Plackett \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e1975\u003c/span\u003e; Luce 1959; Turner et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo analyze shelter selection probabilities while accounting for the sequential choice structure, each ranking was converted into a set of staged choice events: Stage 1 (choice among four shelters), Stage 2 (choice among three remaining shelters), and Stage 3 (choice among two shelters). We fitted conditional logistic regression models with choice-set strata and clustering by individual to account for repeated observations within fish. Two models were considered: m0: chosen\u0026thinsp;~\u0026thinsp;shelter\u0026thinsp;+\u0026thinsp;strata(choice_set) + cluster(individual) and m1: chosen\u0026thinsp;~\u0026thinsp;shelter \u0026times; species\u0026thinsp;+\u0026thinsp;strata(choice_set) + cluster(individual). Odds ratios (OR) were estimated with 95% confidence intervals. Rock was treated as the reference shelter for the baseline species (\u003cem\u003eI. balsanus\u003c/em\u003e).\u003c/p\u003e \u003cp\u003eFinally, to test whether shelter preferences differed between species, we compared models \u003cem\u003em0\u003c/em\u003e and \u003cem\u003em1\u003c/em\u003e using a likelihood ratio test (LRT). For this comparison, models were refitted without the clustering term to obtain comparable likelihood values.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eAll 36 individuals produced complete shelter rankings (ranks 1\u0026ndash;4), and there were no missing values in species identity, individual identity, shelter choice, total length, or body mass. Total length did not differ between species (mean\u0026thinsp;=\u0026thinsp;18.92\u0026thinsp;\u0026plusmn;\u0026thinsp;1.48 cm; Welch\u0026rsquo;s t\u0026thinsp;=\u0026thinsp;0.325, df\u0026thinsp;=\u0026thinsp;33.72, p\u0026thinsp;=\u0026thinsp;0.747). Body mass also did not differ significantly between species (mean\u0026thinsp;=\u0026thinsp;41.87\u0026thinsp;\u0026plusmn;\u0026thinsp;8.31 g; Welch\u0026rsquo;s t\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.638, df\u0026thinsp;=\u0026thinsp;27.30, p\u0026thinsp;=\u0026thinsp;0.529). Given the absence of body size differences, body size was not included as a covariate in subsequent analyses.\u003c/p\u003e \u003cp\u003eThe proportion of individuals ranking each shelter type as their first choice is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. First-choice frequencies did not differ significantly between species (χ\u0026sup2; = 2.333, df\u0026thinsp;=\u0026thinsp;3, p\u0026thinsp;=\u0026thinsp;0.506; Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In both species, Wood was the most frequently selected shelter as first choice (10 of 18 individuals in each species; 55.6%), followed by Rock (27.8% in \u003cem\u003eI. balsanus\u003c/em\u003e and 38.9% in \u003cem\u003eI. punctatus\u003c/em\u003e). Algae was occasionally ranked first by \u003cem\u003eI. balsanus\u003c/em\u003e (11.1%) but not by \u003cem\u003eI. punctatus\u003c/em\u003e, whereas PVC was rarely selected as the first choice in either species (5.6% in both cases).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFirst-choice (rank\u0026thinsp;=\u0026thinsp;1) shelter frequencies and proportions by species.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpecies\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRock n (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWood n (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAlgae n (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePVC n (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eI. balsanus\u003c/em\u003e\u0026nbsp;(n\u0026thinsp;=\u0026thinsp;18)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5 (27.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10 (55.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2 (11.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1 (5.6)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eI. punctatus\u003c/em\u003e\u0026nbsp;(n\u0026thinsp;=\u0026thinsp;18)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7 (38.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10 (55.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0 (0.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1 (5.6)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe complete ranking data also showed a clear overall hierarchy of shelter preference. The Plackett\u0026ndash;Luce model fitted to rankings pooled across species assigned the highest worth to Wood (0.552), followed by Rock (0.320), Algae (0.084), and PVC (0.043). Thus, Wood and Rock together accounted for most of the overall preference weight, whereas Algae and especially PVC were consistently less preferred.\u003c/p\u003e \u003cp\u003eConditional-choice analyses also supported a strong effect of shelter type on selection probability (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). For the baseline species, \u003cem\u003eI. balsanus\u003c/em\u003e, Wood showed higher odds of being selected than Rock (OR\u0026thinsp;=\u0026thinsp;2.004, 95% CI\u0026thinsp;=\u0026thinsp;0.791\u0026ndash;5.076, p\u0026thinsp;=\u0026thinsp;0.143), whereas Algae and PVC showed lower odds of selection than Rock (Algae: OR\u0026thinsp;=\u0026thinsp;0.486, 95% CI\u0026thinsp;=\u0026thinsp;0.212\u0026ndash;1.113, p\u0026thinsp;=\u0026thinsp;0.088; PVC: OR\u0026thinsp;=\u0026thinsp;0.228, 95% CI\u0026thinsp;=\u0026thinsp;0.098\u0026ndash;0.530, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). These results are consistent with the ranking analysis in indicating a preference hierarchy of Wood\u0026thinsp;\u0026gt;\u0026thinsp;Rock\u0026thinsp;\u0026gt;\u0026thinsp;Algae\u0026thinsp;\u0026gt;\u0026thinsp;PVC.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eConditional-choice model results: odds ratios (OR) with 95% confidence intervals for shelter selection, including species-specific contrasts; Rock is the reference shelter for the baseline species (\u003cem\u003eI. balsanus\u003c/em\u003e)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTerm\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWood\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.791\u0026ndash;5.076\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.143\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlgae\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.212\u0026ndash;1.113\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0879\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePVC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.228\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.098\u0026ndash;0.530\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000599\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRock \u0026times;\u0026nbsp;\u003cem\u003eI. punctatus\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.964\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.743\u0026ndash;33.172\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0983\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWood \u0026times;\u0026nbsp;\u003cem\u003eI. punctatus\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.587\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.514\u0026ndash;25.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.197\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlgae \u0026times;\u0026nbsp;\u003cem\u003eI. punctatus\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.918\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.228\u0026ndash;3.699\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.904\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePVC \u0026times;\u0026nbsp;\u003cem\u003eI. punctatus\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003ePVC \u0026times; \u003cem\u003eI. punctatus\u003c/em\u003e was not estimable (NA), consistent with quasi-complete separation due to extremely rare selection of PVC by \u003cem\u003eI. punctatus\u003c/em\u003e within the staged choice sets.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAllowing shelter preferences to vary between species did not significantly improve model fit compared to a model assuming a shared preference structure (likelihood-ratio test: χ\u0026sup2; = 5.822, df\u0026thinsp;=\u0026thinsp;3, p\u0026thinsp;=\u0026thinsp;0.121). Thus, there was no statistically robust evidence that \u003cem\u003eI. balsanus\u003c/em\u003e and \u003cem\u003eI. punctatus\u003c/em\u003e differed in their overall hierarchy of shelter selection. Species-specific coefficients were generally imprecisely estimated, with wide confidence intervals (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In addition, the PVC \u0026times; \u003cem\u003eI. punctatus\u003c/em\u003e contrast was not estimable, consistent with quasi-complete separation caused by the extremely rare selection of PVC by \u003cem\u003eI. punctatus\u003c/em\u003e in the staged choice sets.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eJuvenile \u003cem\u003eIctalurus balsanus\u003c/em\u003e and \u003cem\u003eI. punctatus\u003c/em\u003e exhibited a clear and consistent hierarchy of shelter preference, with Wood and Rock being selected more often than Algae and PVC. This pattern was evident across all analytical approaches used in this study, including first-choice frequencies, the Plackett\u0026ndash;Luce ranking model, and the conditional-choice analysis. Preference tests are widely used in behavioral ecology to infer how animals value alternative options, and combining rank-based inference with staged choice models provides a particularly strong basis for concluding that selection was structured rather than random (Kirkden and Pajor \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eShelter preference is expected to have ecological consequences because refuges can mediate key processes that determine survival and local distribution. The refuge concept links sheltering behavior to predator avoidance, resource limitation, and the outcomes of species interactions (Berryman and Hawkins \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). In fishes and other aquatic animals, sheltering can reduce exposure to predators and environmental stress, and shelter use can also be shaped by external conditions (e.g., seasonal constraints or light-related risk), making refuge selection a plausible axis of the spatial niche (Valdimarsson and Metcalfe \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e1998\u003c/span\u003e; Kerry and Bellwood \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Under this framework, the strong skew toward Wood and Rock suggests that these shelter types likely provide functional benefits that the fish can detect behaviorally, even though the adaptive value of specific shelters often requires direct tests that link choice to performance (Arce and Alcaraz \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2012\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this study, invasive \u003cem\u003eI. punctatus\u003c/em\u003e did not show a shift in shelter preference relative to the endemic \u003cem\u003eI. balsanus\u003c/em\u003e. Both species exhibited broadly similar preference hierarchies, and allowing shelter preferences to vary between species did not significantly improve model fit. This implies substantial microhabitat overlap in refuge use where the two species coexist, and it reinforces concerns about competitive interactions in invaded systems given that \u003cem\u003eI. punctatus\u003c/em\u003e is established in tributaries of the Balsas basin and is considered a conservation concern for native fishes (Mej\u0026iacute;a-Mojica et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Although, \u0026ldquo;shared preference\u0026rdquo; does not necessarily mean \u0026ldquo;no impact\u0026rdquo;, in other invaded systems, competition for access to preferred shelters can occur even when species share similar habitat requirements, and invasive species can restrict native access to shelters via interference or displacement (Van Kessel et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Champneys et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This inference is consistent with prior evidence that juvenile \u003cem\u003eI. punctatus\u003c/em\u003e can alter the diel activity pattern of juvenile \u003cem\u003eI. balsanus\u003c/em\u003e, indicating that behavioral interactions between the two species can be strong even in controlled conditions (Rubio-Plascencia et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAcross aquatic taxa, strong shelter preferences are commonly linked to specific physical properties of refuges, which provides a mechanistic context for our Wood/Rock versus PVC pattern. For example, spiny lobsters can strongly prefer shelters that provide shade or cover and appropriate entrance configurations, and their den choice can vary with predation risk and social conditions (Spanier and Zimmer-Faust \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e1988\u003c/span\u003e; Eggleston and Lipcius 1992). Similarly, slipper lobsters prefer shaded or opaque shelters over transparent ones, consistent with concealment-based mechanisms (Spanier and Almog-Shtayer \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e1992\u003c/span\u003e). In crabs, refuge geometry can influence both preference and agonistic interactions; swimming crabs show significant shelter-shape preferences, and some configurations can reduce willingness to fight (Zhang et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These cross-taxa examples support the idea that animals evaluate shelters based on how refuge design alters concealment and escape options.\u003c/p\u003e \u003cp\u003ePVC was particularly rarely selected in our trials, a result that aligns with multiple studies showing that smooth artificial shelters may be less preferred than rougher or more complex alternatives. In Arctic charr aquaculture, a PVC-based shelter design with grooves altered behavior and performance, highlighting that fine-scale structural features of artificial refuges can matter (Benha\u0026iuml;m et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). In the zebra pleco, shelter selection in rearing conditions shows that fish can use PVC, but may show higher utilization of alternative, rougher shelters (like brick or clay) relative to smooth PVC pipes (Ramos et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Outside fishes, sea cucumbers also show shelter preferences depending on material and structural design (e.g., PVC versus cement-type materials and opening characteristics), emphasizing that material effects are widespread across aquatic animals (Dong et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Kwon et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). It is possible that surface texture, internal complexity, and perceived concealment contribute to why Wood and Rock were preferred over PVC in our catfishes.\u003c/p\u003e \u003cp\u003eIn addition to geometry and texture, visual and camouflage-related cues provide another plausible explanation for the avoidance of PVC in particular. Camouflage theory predicts that matching background color and reducing conspicuousness can reduce detection risk, and several studies show that animals can exhibit color-associated preferences in sheltering contexts (Stevens and Merilaita 2008; Pessani and Tirelli \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). For example, mud crabs show shelter color preferences in controlled tests (Kawamura et al. 2020), and hermit crab shell choice can be related to chromatic resemblance between crab and occupied shell (Pessani and Tirelli \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). In this study, it is plausible that the darker, heterogeneous appearance of Wood (and, depending on construction, Rock) may offer better concealment than relatively uniform, light-colored PVC, especially under daylight conditions.\u003c/p\u003e \u003cp\u003eShelter preference can also be influenced by trophic or chemical pathways, because some refuges accumulate biofilms or concentrate prey resources. In rivers, in-stream wood can function not only as physical cover but also as a substrate linked to food resources and habitat patchiness, making wood a biologically \u0026ldquo;active\u0026rdquo; refuge type (Pettit et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). In aquaculture contexts, sea cucumbers show differential responses to shelter materials and colors, and shelter design is often evaluated precisely because it changes local conditions experienced by the organism (Dong et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Kwon et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In our study, refuges were cleaned and treated to reduce odor absorption and chemical cue bias, so we do not interpret our results as driven by biofilm accumulation. However, in natural settings, chemical and trophic differences among refuge types could reinforce structural preferences and further increase overlap between native and invasive catfishes on the same refuge patches.\u003c/p\u003e \u003cp\u003eFrom a management perspective, many restoration programs aim to increase physical complexity by maintaining rocky substrate and large wood inputs, because structural complexity is linked to habitat quality and behavioral opportunities for aquatic organisms (Tokeshi and Arakaki \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Pettit et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Our results suggest that such actions would likely benefit both catfish species rather than selectively favoring the endemic fish, since both \u003cem\u003eI. balsanus\u003c/em\u003e and invasive \u003cem\u003eI. punctatus\u003c/em\u003e preferred the same refuge types. In addition, PVC and other artificial shelters are widely used in aquaculture and in applied contexts to mitigate aggression or cannibalism, and there is active work on screening shelter designs for different species (Benha\u0026iuml;m et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). However, our results caution against assuming that PVC is an appropriate refuge substitute for all fishes, because preference can be highly species- and context-dependent and may fail to match key physical or sensory features of natural refuges.\u003c/p\u003e \u003cp\u003eIt is worthy to note that our trials were conducted in isolation and therefore did not include direct interspecific interactions; the realized use of shelters under interference competition may differ from preference expressed without competitors. This matters because shelter competition is a documented mechanism of invasive impacts in both fishes and crustaceans, often producing exclusion from preferred shelters or dominance by one species (Van Kessel et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Champneys et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Vorburger and Ribi \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). Additionally, some shelter\u0026ndash;species combinations were rare (e.g., PVC by \u003cem\u003eI. punctatus\u003c/em\u003e), producing quasi-complete separation and limiting precision for some interaction terms, although not altering the overall hierarchy supported by multiple methods. We also used juveniles of both species; because sheltering behavior and refuge needs can change across ontogeny in some taxa, future work should test whether the same hierarchy holds across life stages. Finally, preference may shift with predation risk and social context, as demonstrated in classic shelter-selection work on lobsters (Eggleston and Lipcius 1992), suggesting that predator-cue experiments and direct competition trials are the most direct way to test the adaptive and ecological consequences of shelter choice in this system.\u003c/p\u003e"},{"header":"Statements and Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT, IA-202424). We thank Olivia de los Santos, Ignacio Morales-Salas, Maria Pérez-Cruz and Cassandra Rubio for their technical assistance in the field.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003cbr\u003e\u003c/strong\u003eThis study was funded by the Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT, IA-202424).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003cbr\u003e\u003c/strong\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003cbr\u003e\u003c/strong\u003eAll authors conceived the study. HMJ and SAR conducted experiments. LMB analysed the data. LMB wrote the first draft. All authors revised the manuscript and approved the final version.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of generative AI\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring the preparation of this work, the authors used ChatGPT (OpenAI) to assist with English language checking. After using this tool, the authors reviewed the content and take full responsibility for the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003cbr\u003e\u003c/strong\u003eThe study was approved by the Commission of Ethics and Scientific Responsibility, Faculty of Sciences, UNAM (CEARC/19062023 PI_06_06_2023_Burciaga). All experimental procedures in this study were performed complying with Mexican legislation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData and Code Availability\u003cbr\u003e\u003c/strong\u003eData and code are available at figshare (https://doi.org/10.6084/m9.figshare.31724407)\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eArce E, Alcaraz G (2012) Shell preference in a hermit crab: comparison between a matrix of paired comparisons and a multiple-alternative experiment. Mar Biol 159:853\u0026ndash;862. https://doi.org/10.1007/s00227-011-1861-x\u003c/li\u003e\n\u003cli\u003eBenha\u0026iuml;m D, Leblanc CA, Lucas G (2009) Impact of a new artificial shelter on Arctic charr (\u003cem\u003eSalvelinus alpinus\u003c/em\u003e L.) behaviour and culture performance during the endogenous feeding period. Aquaculture 295:38\u0026ndash;43. https://doi.org/10.1016/j.aquaculture.2009.06.024\u003c/li\u003e\n\u003cli\u003eBerryman AA, Hawkins BA (2006) The refuge as an integrating concept in ecology and evolution. Oikos 115:192\u0026ndash;196. https://doi.org/10.1111/j.0030-1299.2006.15188.x\u003c/li\u003e\n\u003cli\u003eBlanchet S, Loot G, Bernatchez L, Dodson JJ (2008) The effects of abiotic factors and intraspecific versus interspecific competition on the diel activity patterns of Atlantic salmon (\u003cem\u003eSalmo salar\u003c/em\u003e) fry. Can J Fish Aquat Sci 65:1545\u0026ndash;1553. https://doi.org/10.1139/F08-079\u003c/li\u003e\n\u003cli\u003eChampneys T, Genner MJ, Ioannou CC (2021) Invasive Nile tilapia dominates a threatened indigenous tilapia in competition over shelter. Hydrobiologia 848:3747\u0026ndash;3762. https://doi.org/10.1007/s10750-020-04341-8\u003c/li\u003e\n\u003cli\u003eCohen AN, Webb SK (eds) (2002) Biological invasions in aquatic ecosystems: impacts on restoration and potential for control. San Francisco Estuary Institute, Oakland\u003c/li\u003e\n\u003cli\u003eCucherousset J, Olden JD (2011) Ecological impacts of non-native freshwater fishes. Fisheries 36:215\u0026ndash;230. https://doi.org/10.1080/03632415.2011.574578\u003c/li\u003e\n\u003cli\u003eDong G, Dong S, Wang F, Tian X (2010) Effects of materials, incubation time and colors of artificial shelters on behavior of juvenile sea cucumber \u003cem\u003eApostichopus japonicus\u003c/em\u003e. Aquacult Eng 43:1\u0026ndash;5. https://doi.org/10.1016/j.aquaeng.2010.01.002\u003c/li\u003e\n\u003cli\u003eKerry JT, Bellwood DR (2017) Environmental drivers of sheltering behaviour in large reef fishes. Mar Pollut Bull 125:254\u0026ndash;259. https://doi.org/10.1016/j.marpolbul.2017.08.036\u003c/li\u003e\n\u003cli\u003eKirkden RD, Pajor EA (2006) Using preference, motivation and aversion tests to ask scientific questions about animals\u0026rsquo; feelings. Appl Anim Behav Sci 100:29\u0026ndash;47. https://doi.org/10.1016/j.applanim.2006.04.009\u003c/li\u003e\n\u003cli\u003eKronfeld-Schor N, Dayan T (2003) Partitioning of time as an ecological resource. Annu Rev Ecol Evol Syst 34:153\u0026ndash;181. https://doi.org/10.1146/annurev.ecolsys.34.011802.132435\u003c/li\u003e\n\u003cli\u003eKwon I, Lee K, Kim T (2019) Shelter material and shape preferences of the sea cucumber \u003cem\u003eApostichopus japonicus\u003c/em\u003e. Aquaculture 508:206\u0026ndash;213. https://doi.org/10.1016/j.aquaculture.2019.05.011\u003c/li\u003e\n\u003cli\u003eLara-Rivera AL, Parra-Bracamonte GM, Sifuentes-Rinc\u0026oacute;n AM, Goj\u0026oacute;n-B\u0026aacute;ez HH, Rodr\u0026iacute;guez-Gonz\u0026aacute;lez H, Montelongo-Alfaro IO (2015) El bagre de canal (\u003cem\u003eIctalurus punctatus\u003c/em\u003e Rafinesque, 1818): estado actual y problem\u0026aacute;tica en M\u0026eacute;xico. Lat Am J Aquat Res 43:424\u0026ndash;434. https://doi.org/10.3856/vol43-issue3-fulltext-4\u003c/li\u003e\n\u003cli\u003eMej\u0026iacute;a-Mojica H, Paredes-Lira ME, Beltr\u0026aacute;n-L\u0026oacute;pez RG (2013) Primer registro y establecimiento del bagre de canal \u003cem\u003eIctalurus punctatus\u003c/em\u003e (Siluriformes: Ictaluridae) en un tributario del R\u0026iacute;o Balsas, M\u0026eacute;xico. Hidrobiol\u0026oacute;gica 23:456\u0026ndash;459\u003c/li\u003e\n\u003cli\u003eMillidine KJ, Armstrong JD, Metcalfe NB (2006) Presence of shelter reduces maintenance metabolism of juvenile salmon. Funct Ecol 20:839\u0026ndash;845. https://doi.org/10.1111/j.1365-2435.2006.01166.x\u003c/li\u003e\n\u003cli\u003ePosit Team (2024) RStudio: Integrated Development Environment for R. Posit Software, Boston\u003c/li\u003e\n\u003cli\u003ePettit NE, Warfe DM, Kennard MJ, Pusey BJ, Davies PM, Douglas MM (2013) Dynamics of in-stream wood and its importance as fish habitat in a large tropical floodplain river. River Res Appl 29:864\u0026ndash;875\u003c/li\u003e\n\u003cli\u003ePlackett RL (1975) The analysis of permutations. J R Stat Soc Ser C Appl Stat 24:193\u0026ndash;202. https://doi.org/10.2307/2346567\u003c/li\u003e\n\u003cli\u003ePessani D, Tirelli T (2006) Chromatic patterns of the hermit crab \u003cem\u003eCalcinus tubularis\u003c/em\u003e related to the occupied shell. Hydrobiologia 557:107\u0026ndash;112. https://doi.org/10.1007/s10750-005-1313-z\u003c/li\u003e\n\u003cli\u003eR Core Team (2024) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna\u003c/li\u003e\n\u003cli\u003eRamos FM, Recuero LB, Silva TVN, Fujimoto RY, Lee JT, Ferreira-Torres M (2013) Shelter selection in the Amazonian zebra pleco \u003cem\u003eHypancistrus zebra\u003c/em\u003e. J Appl Ichthyol 29:920\u0026ndash;927. https://doi.org/10.1111/jai.12176\u003c/li\u003e\n\u003cli\u003eRubio-Plascencia C, Kiere LM, Burciaga LM (2025) Juveniles of the endemic catfish \u003cem\u003eIctalurus balsanus\u003c/em\u003e are less nocturnal in the presence of juvenile \u003cem\u003eIctalurus punctatus\u003c/em\u003e. Environ Biol Fishes 108:957\u0026ndash;968. https://doi.org/10.1007/s10641-025-01700-z\u003c/li\u003e\n\u003cli\u003eSchoener TW (1974) Resource partitioning in ecological communities. Science 185:27\u0026ndash;39. https://doi.org/10.1126/science.185.4145.27\u003c/li\u003e\n\u003cli\u003eSmith TM, Smith RL (2009) Elements of ecology, 7th edn. Pearson Benjamin Cummings, San Francisco\u003c/li\u003e\n\u003cli\u003eSpanier E, Zimmer-Faust RK (1988) Some physical properties of shelter that influence den preference in spiny lobsters. J Exp Mar Biol Ecol 121:137\u0026ndash;149. https://doi.org/10.1016/0022-0981(88)90251-1\u003c/li\u003e\n\u003cli\u003eSpanier E, Almog-Shtayer G (1992) Shelter preferences in the Mediterranean slipper lobster: effects of physical properties. J Exp Mar Biol Ecol 164:103\u0026ndash;116. https://doi.org/10.1016/0022-0981(92)90139-2\u003c/li\u003e\n\u003cli\u003eStevens M, Merilaita S (2009) Animal camouflage: current issues and new perspectives. Philos Trans R Soc B 364:423\u0026ndash;427. https://doi.org/10.1098/rstb.2008.0217\u003c/li\u003e\n\u003cli\u003eTokeshi M, Arakaki S (2012) Habitat complexity in aquatic systems: fractals and beyond. Hydrobiologia 685:27\u0026ndash;47. https://doi.org/10.1007/s10750-011-0832-z\u003c/li\u003e\n\u003cli\u003eTurner HL, van Etten J, Firth D, Kosmidis I (2020) Modelling rankings in R: the PlackettLuce package. Comput Stat 35:1027\u0026ndash;1057. https://doi.org/10.1007/s00180-020-00959-3\u003c/li\u003e\n\u003cli\u003eVacher C, Tamaddoni-Nezhad A, Kamenova S et al (2016) Learning ecological networks from next-generation sequencing data. In: Woodward G, Bohan DA (eds) Adv Ecol Res, vol 54. Academic Press, Oxford, pp 1\u0026ndash;39. https://doi.org/10.1016/bs.aecr.2015.10.004\u003c/li\u003e\n\u003cli\u003eValdimarsson SK, Metcalfe NB (1998) Shelter selection in juvenile Atlantic salmon, or why do salmon seek shelter in winter? J Fish Biol 52:42\u0026ndash;49. https://doi.org/10.1111/j.1095-8649.1998.tb01551.x\u003c/li\u003e\n\u003cli\u003evan Kessel N, Dorenbosch M, de Boer MRM, Leuven RSEW, van der Velde G (2011) Competition for shelter between four invasive gobiids and two native benthic fish species. Curr Zool 57:844\u0026ndash;851. https://doi.org/10.1093/czoolo/57.6.844\u003c/li\u003e\n\u003cli\u003eVorburger C, Ribi G (1999) Aggression and competition for shelter between a native and an introduced crayfish in Europe. Freshw Biol 42:111\u0026ndash;119. https://doi.org/10.1046/j.1365-2427.1999.00465.x\u003c/li\u003e\n\u003cli\u003eZhang H, Zhu B, Yu L, Liu D, Wang F, Lu Y (2021) Selection of shelter shape by swimming crab (\u003cem\u003ePortunus trituberculatus\u003c/em\u003e). Aquac Rep 21:100908. https://doi.org/10.1016/j.aqrep.2021.100908\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"biological-invasions","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"binv","sideBox":"Learn more about [Biological Invasions](https://www.springer.com/journal/10530)","snPcode":"10530","submissionUrl":"https://submission.nature.com/new-submission/10530/3","title":"Biological Invasions","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"biological invasions, habitat selection, shelter use, conditional logistic regression","lastPublishedDoi":"10.21203/rs.3.rs-9124421/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9124421/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eShelters are an important component of the spatial niche of many freshwater fishes, providing protection from predators and environmental stressors and influencing patterns of habitat use. When native and invasive species rely on similar refuges, overlap in shelter use may promote competition and affect the structure of aquatic communities. Here, we compared the shelter preferences of juvenile individuals of the endemic Balsas catfish \u003cem\u003eIctalurus balsanus\u003c/em\u003e and the invasive channel catfish \u003cem\u003eI. punctatus\u003c/em\u003e. Each individual was presented with four shelter types (Rock, Wood, Algae, and PVC) in a sequential rank-choice experiment that generated a preference hierarchy. Both species exhibited a consistent preference structure. Wood was most frequently selected as the first refuge, followed by Rock, whereas Algae and especially PVC were rarely chosen. Analyses based on full ranking data likewise revealed a strong preference hierarchy (Wood\u0026thinsp;\u0026gt;\u0026thinsp;Rock\u0026thinsp;\u0026gt;\u0026thinsp;Algae\u0026thinsp;\u0026gt;\u0026thinsp;PVC). Despite the invasive status of \u003cem\u003eI. punctatus\u003c/em\u003e, shelter preference patterns were broadly similar between species. These results indicate substantial overlap in refuge use between the endemic and invasive catfishes. Because structural refuges can influence survival, resting behavior, and exposure to predators, such overlap could promote competition when preferred shelters are limited in natural habitats. Understanding how native and invasive species use refuges may therefore help clarify potential mechanisms of interaction and inform management strategies in invaded freshwater ecosystems.\u003c/p\u003e","manuscriptTitle":"Shared shelter preferences in an endemic and an invasive catfish (Ictalurus balsanus and Ictalurus punctatus)","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-08 07:27:23","doi":"10.21203/rs.3.rs-9124421/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2026-05-09T14:33:29+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-02T09:32:48+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Biological Invasions","date":"2026-03-19T21:08:53+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-16T06:37:15+00:00","index":"","fulltext":""},{"type":"submitted","content":"Biological Invasions","date":"2026-03-14T14:21:07+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"biological-invasions","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"binv","sideBox":"Learn more about [Biological Invasions](https://www.springer.com/journal/10530)","snPcode":"10530","submissionUrl":"https://submission.nature.com/new-submission/10530/3","title":"Biological Invasions","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"7c6b1a3a-92fa-4de8-9e11-d93611369db7","owner":[],"postedDate":"April 8th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"","date":"2026-05-09T14:33:29+00:00","index":0,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-08T07:27:23+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-08 07:27:23","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9124421","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9124421","identity":"rs-9124421","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}