Does donor group size matter? The response of guppies (Poecilia reticulata) and convict cichlids (Amatitlania nigrofasciata) to disturbance cues from conspecific and heterospecific donors

Abstract

Prey are under immense pressure to make context-specific, behavioural decisions. Prey use public information to reduce the costs associated with making inappropriate decisions. Chemical cues are commonly used by aquatic vertebrates to assess local threats and facilitate behavioural decision making. Previous studies on chemosensory assessment of risk have largely focused on damage-released alarm cues, with the cues released by disturbed or stressed prey (i.e., disturbance cues) receiving less attention. Disturbance cues are “early-warning signals” common among aquatic vertebrates that may warn conspecific and heterospecific prey guild members of potential risk. Initially, we conducted a series of laboratory studies to determine (i) if guppies (Poecilia reticulata Peters, 1859) produce and respond to disturbance cues and (ii) if relative concentration (donor group size) determines response intensity. Secondly, we examined if guppies and convict cichlids (Amatitlania nigrofasciata (Günther, 1867)) show similar response patterns to their own vs. heterospecific disturbance cues. Our results suggest that guppies exhibit increased predator avoidance behaviour to conspecific disturbance cues (relative to water from undisturbed conspecifics) and increased donor group size lead to stronger antipredator responses. However, although guppies and cichlids respond to each other’s disturbance cues, we found no effect of donor group size towards heterospecific disturbance cues. Our results suggest that disturbance cues are not generalized cues and present a degree of species-specificity.

Disturbance cues as a source of risk assessment information under natural conditions

“…Laboratory studies of the presence and function of disturbance cues have typically included the odour of undisturbed conspecifics as a control treatment (Bairos‐Novak et al, ; Goldman et al, ; Vavrek et al, ). We did not include this as a control in our field trials for two reasons.…”

“…Disturbance cues were generated blind to population. This method has been shown to be a reliable way to collect disturbance cues (Goldman et al, ; Vavrek & Brown, ). We froze the disturbance cues in 20‐ml aliquots at −20°C until used in the field.…”

“…Disturbance cues are released by stressed or disturbed prey (Abreu et al, ) and thought to be expelled across the gill epithelium or in the urine of aquatic prey (Hazlett, ; Wisenden et al, ). These cues provide information to both conspecific and heterospecific prey guild members (Goldman, Singh, Demers, Feyten, & Brown, ) and have been shown to be widespread across a variety of taxa (Bairos‐Novak, Mitchell, Crane, Chivers, & Ferrari, ). The early‐warning nature of these cues is beneficial to prey, as they afford the opportunity to engage in antipredator behaviours at the earliest stages of an encounter with a predator (Goldman et al, ; Wisenden et al, ).…”

“…These cues provide information to both conspecific and heterospecific prey guild members (Goldman, Singh, Demers, Feyten, & Brown, ) and have been shown to be widespread across a variety of taxa (Bairos‐Novak, Mitchell, Crane, Chivers, & Ferrari, ). The early‐warning nature of these cues is beneficial to prey, as they afford the opportunity to engage in antipredator behaviours at the earliest stages of an encounter with a predator (Goldman et al, ; Wisenden et al, ). Additionally, these cues strengthen predator recognition learning (Ferrari et al, ), and have been shown to increase survival rates during staged predator encounters (Mirza & Chivers, ).…”

“…Additionally, these cues strengthen predator recognition learning (Ferrari et al, ), and have been shown to increase survival rates during staged predator encounters (Mirza & Chivers, ). While a growing number of studies have shown that disturbance cues function as sources of risk assessment information, these studies have been restricted to laboratory and semi‐natural conditions (Abreu et al, ; Bairos‐Novak et al, ; Ferrari et al, ; Goldman et al, ; Hazlett, ; Mirza & Chivers, ; Wisenden, ; Wisenden et al, ). Therefore, given the preponderance of laboratory studies, field studies are needed to verify that these cues function under ecologically relevant conditions (i.e.…”