Similar behavioral but different endocrine responses to conspecific interactions in hand-raised wolves and dogs

Abstract

Domestication has altered dogs' conspecific social organization compared to their closest, non-domesticated relatives, gray wolves. Wolves live in packs whose survival depends on coordinated behavior, but dogs rely less on conspecifics, which predicts greater cohesiveness in wolf than dog packs. Endocrine correlates such as oxytocin and glucocorticoids modulate group cohesion resulting in species-specific differences in social interactions. We found that although wolves' and dogs' observable behavioral reactions to a territorial threat and separation from the pack were similar, hormonal responses differed. Wolves' but not dogs' oxytocin and glucocorticoid concentrations correlated positively with territorial behaviors and only wolves showed increased glucocorticoid concentrations after separation from their pack. Together, results suggest stronger emotional activation to threats to group integrity in wolves than dogs, in line with their socio-ecology.

Keywords: Canine behavior; Canine physiology.

Graphical abstract

Figure 1

Overview of conditions (A–H) Experimental and control conditions in wolves and dogs. Induced chorus howling (A: wolves, E: dogs); mock territorial patrol (B: wolves F: dogs); separation from pack (C: wolf; G: dog); undisturbed period in familiar environment with pack mates present/control condition (D: wolves, H: dogs).

Figure 2

Oxytocin and synchronized movement (A and B) Association of synchronized locomotion and urinary OTM concentrations (pg/mL, corrected for specific gravity, SG) in a) dogs (N = 10 individuals) and b) wolves (N = 9 individuals). The dashed lines represent the fitted model for the effect of synchronized locomotion on urinary OTM concentrations, given all control predictors and random factors. Each dot represents a sample. Each animal provided 1–3 samples (the model accounted for repeated sampling of the same individuals).

Figure 3

Oxytocin and territorial behavior (A and B) Association of territorial behavior and urinary OTM concentrations (pg/mL, corrected for specific gravity, SG) in (A) dogs (N = 10 individuals) and (B) wolves (N = 9 individuals). The dashed lines represent the fitted model for the effect of territorial behavior on urinary OTM concentrations, given all control predictors and random factors. Each dot represents a sample. Each animal provided 1–3 samples (the model accounted for repeated sampling of the same individuals).

Figure 4

Glucocorticoids across conditions Urinary GCM concentrations (ng/mL SG) of wolves (N = 10; white boxes) and dogs (N = 10; light gray boxes) across test conditions Indicated are medians and quartiles (horizontal lines with boxes) as well as the fitted model and its 95% confidence intervals (thick horizontal lines with error bars). Gray dots represent individual samples. ∗∗p ≤ 0.01, ∗p ≤ 0.05.

Figure 5

Glucocorticoids and territorial behavior (A and B) Association of territorial behavior and urinary GCM concentrations (ng/mL, corrected for specific gravity, SG) in a) dogs (N = 10 individuals) and b) wolves (N = 9 individuals). The dashed lines represent the fitted model for the effect of territorial behavior on urinary GCM concentrations, given all control predictors and random factors. Each dot represents a sample. Each animal provided 1–3 samples (the model accounted for repeated sampling of the same individuals).

Figure 6

Glucocorticoids and solo howling Association of solo howling and urinary GCM concentrations (ng/mL, corrected for specific gravity, SG) The dashed lines represent the fitted model for the effect of solo howling on urinary GCM concentrations, given all control predictors and random factors. Each dot represents a sample. Each animal provided 1–3 samples (the model accounted for repeated sampling of the same individuals).

Figure 7

Link between oxytocin and glucocorticoids (control condition) (A and B) Effect of urinary OTM on urinary GCM concentrations (ng/mL, corrected for specific gravity, SG) in a) dogs (N = 10 individuals) and b) wolves (N = 9 individuals). The dashed line represents the fitted model of the effect of urinary OTM on urinary GCM concentrations following the control condition (baseline samples).

Figure 8

Link between oxytocin and glucocorticoids (chorus howling condition) Effect of urinary OTM on urinary GCM concentrations (ng/mL, corrected for specific gravity, SG) The dotted line represents the fitted model of the effect of urinary OTM on urinary GCM concentrations following the induced chorus howling condition in both dogs and wolves.