The threat-sensitive predator avoidance model predicts that prey should balance the intensity of antipredator responses against perceived predation risk, resulting in a graded response pattern. Recent studies have demonstrated considerable interspecific variation in the intensity of threat-sensitive response patterns, ranging from strongly graded to relatively nongraded or "hypersensitive" threat-sensitive response patterns. Here, we test for intraspecific plasticity in threat-sensitive responses by varying group size. We exposed juvenile convict cichlids, Archocentrus nigrofasciatus (Günther, 1867), as individuals or in small (groups of three) or large (groups of six) shoals to a series of dilutions of conspecific chemical alarm cues and a distilled water control. Singleton cichlids exhibited significant reductions in time spent moving and in frequency of foraging attempts (relative to distilled water controls) when exposed to a 12.5% dilution of conspecific alarm cue, with no difference in response intensity at higher stimulus concentrations, suggesting a nongraded (hypersensitive) response pattern. Small shoals exhibited a similar response pattern, but at a higher response threshold (25% dilution of stock alarm cue solution). Large shoals, however, exhibited a graded response pattern. These results suggest that group size influences the trade-off between predator avoidance and other fitness related activities, resulting in flexible threat-sensitive response patterns.

Many prey fishes rely on damage-released chemical alarm cues to detect and avoid predators. The ability to use these cues has been shown to confer considerable survival benefits to individuals. While several laboratory studies and a single field study have demonstrated that an ambient pH of 6.0 impairs fishes in their ability to detect these alarm cues, no study had yet compared the response to alarm cue exposures across populations residing in multiple streams of a different acidity level. In our study, we conducted live behavioural observations in five nursery streams, ranging in pH from 5.71 to 7.49 on two age classes (young of the year and parr) of wild juvenile Atlantic salmon (Salmo salar). We aimed to assess if the detection of these chemical alarm cues was constantly dependant on the ambient pH or if variations in the detection occurred among populations of the different streams regardless of the ambient acidity level. Our results demonstrated that salmon present in any acidic stream did not respond to alarm cues, while those in neutral streams exhibited typical alarm responses.Résumé : Plusieurs poissons proies utilisent les signaux d'alerte chimiques libérés par les blessures pour détecter et éviter les prédateurs. Il a été démontré que la capacité d'utiliser ces signaux procure aux individus des bénéfices significatifs de survie. Bien que plusieurs études de laboratoire et une seule étude de terrain aient prouvé qu'un pH ambiant de 6,0 compromet la capacité des poissons à percevoir ces signaux d'alerte, aucune recherche n'a encore comparé les réactions à la présence des signaux d'alerte dans une gamme de populations habitant des cours d'eau avec des niveaux différents d'acidité. Dans notre travail, nous avons mené des observations comportementales en nature dans cinq cours d'eau d'élevage dont les pH variaient de 5,71 à 7,49 chez deux classes d'âge (jeunes de l'année et tacons) de jeunes saumons atlantiques (Salmo salar) sauvages. Nous tentons de déterminer si la perception de ces signaux chimiques d'alerte dépend à tout coup du pH ambiant ou s'il existe des variations entre les populations des différents cours d'eau indépendantes des niveaux ambiants d'acidité. Nos résultats démontrent que tous les saumons qui habitent les cours d'eau acides ne réagissent pas aux signaux d'alerte, alors que ceux des cours d'eau neutres montrent des réactions typiques d'alerte.[Traduit par la Rédaction] Leduc et al. 2363

No abstract

Many organisms rely on chemosensory cues to mediate predation risks. Recent studies have demonstrated impaired chemosensory detection ability under weak acidification. Because rainfall may lead to episodic acidification of surface water, we assessed the effects of acid rain on chemosensory alarm functions. Under natural conditions, we quantified alarm behaviour of juvenile Atlantic salmon (Salmo salar) exposed to conspecific chemical alarm cues before and following rainfall. Before rainfall, salmon were capable of an alarm response in the study streams. After rainfall, salmon from Devil’s Brook did not respond to the alarm cues whereas the detection of salmon from Catamaran Brook (a comparable stream having higher acid neutralising capacity) was maintained. To relate these findings to predator–prey encounters, we performed a second experiment where we staged encounters between prey (rainbow trout, Oncorhynchus mykiss) and predator (largemouth bass, Micropterus salmoides) exposed to acidified and unacidified rainbow trout chemical alarm cues. Trout exposed to acidified alarm cues survived for a significantly shorter amount of time than trout exposed to unacidified alarm cues, whereas no difference in overall predator behaviour was observed. Our results suggest that episodic acidification in small nursery streams may disrupt the chemical information mediated by the chemical alarm cues that can translate into higher survival costs for prey.

Even at sublethal concentrations, various anthropogenic pollutants may disrupt the transfer of chemosensory information, often inducing maladaptive behavioral responses. Recent studies of freshwater prey fishes have shown impaired abilities to respond to damage-released chemical alarm cues from conspecifics under weakly acidic conditions (pH ; 6.0). Several factors acting individually or collectively may account for such chemosensory impairment. By itself, acidification may chemically disrupt the alarm cues and affect fish olfactory functions. Alternatively, differences in local environmental conditions may affect biochemical composition, quantity of chemical alarm cues produced by epidermal tissue, or both, leading to variations in alarm response. Our goal was to assess whether the ability to produce and detect conspecific chemical alarm cues is similar in individuals reared under neutral versus acidic conditions. We conducted two experiments in which we measured the behavioral response of wild juvenile Atlantic salmon Salmo salar exposed to chemical alarm cues. In particular, we looked for differences in the ability of individual fish to (1) produce alarm cues capable of eliciting consistent antipredator behavior in conspecifics and (2) detect alarm cues upon the fish's introduction into a stream with a pH differing from that of the stream of origin; the latter experiment involved reciprocal transplant of fish between neutral (pH range ; 7.0-7.3) and acidic (pH range ; 5.9-6.3) sites. Our results demonstrate that the ability to produce and respond to chemical alarm cues is maintained in Atlantic salmon reared under acidic conditions and did not differ from that of fish reared under neutral conditions. Overall, these data suggest that no permanent olfactory damage occurred under reduced pH and, likewise, no significant difference in functional alarm cue production existed between Atlantic salmon reared under neutral and acidic conditions. Short-term reduction in olfactory sensitivity and degradation of the chemical alarm cues under acidic conditions are the likely mechanisms affecting detection of these important chemicals by prey fish.