Cross-sensory modulation in a future top predator, the young Nile crocodile

Abstract

Animals routinely receive information through different sensory channels, and inputs from a modality may modulate the perception and behavioural reaction to others. In spite of their potential adaptive value, the behavioural correlates of this cross-sensory modulation have been poorly investigated. Due to their predator life, crocodilians deal with decisional conflicts emerging from concurrent stimuli. By testing young Crocodylus niloticus with sounds in the absence or presence of chemical stimuli, we show that (i) the prandial (feeding) state modulates the responsiveness of the animal to a congruent, i.e. food-related olfactory stimulus, (ii) the prandial state alters the responsiveness to an incongruent (independent of food) sound, (iii) fasted, but not sated, crocodiles display selective attention to socially relevant sounds over noise in presence of food odour. Cross-sensory modulation thus appears functional in young Nile crocodiles. It may contribute to decision making in the wild, when juveniles use it to interact acoustically when foraging.

Keywords: acoustic; behaviour; cross-sensory modulation; olfaction; playback; prandial state.

Figure 1.

Experimental arena divided into two zones (1 and 2) and four orientation areas (north, south, east and west). An active loudspeaker was set up on one side, while there was either water or meat on the other side. To avoid spatial bias, both sides were permuted between tested individuals. Cardinal points were used to identify the animal's head orientation.

Figure 2.

Young Nile crocodiles' preference to food odour versus water according to their prandial state. (a) Double-choice testing device (red dot: area with meat; blue dot: area with water). The boxplots show the presence time over each of the areas for (b) fasted (n = 10) and (c) sated (n = 13) crocodiles. n.s., non-significant difference; **p ≤ 0.01.

Figure 3.

Behavioural response of young Nile crocodiles to sound stimuli according to their prandial state (fasted or sated) and the absence or presence of food. (a) The time spent oriented towards the loudspeaker (head orientation) was assessed during both the pre-playback period (silence) and the playback period (calls or noise). (b) The time spent close to the loudspeaker (LS) and the time spent at the opposite side (containing either water (W) or food (F)) were measured during the playback of sound signals (calls or noise). Beanplots (‘beanplot’ R package) combine individual observations (white horizontal lines), dataset distribution, the overall dataset average (dashed horizontal line) and the average for each subset (black horizontal lines). n.s., non-significant, ^†p < 0.1, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.