Animals in nature seldom use their maximum performance abilities, likely in part due to context-dependent differences in performance motivation. Despite interest in the factors affecting performance expression, the physiological mechanisms underlying variation in performance motivation are poorly understood. We manipulated levels of the biogenic amine octopamine (OA) to test the hypothesis that OA drives motivation to express maximum bite force in male house crickets. We also tested the effect of antenna removal on bite force given prior evidence of potential links among antennaectomy, aggression, and OA. We found that administration of an OA antagonist, epinastine, significantly decreases realized maximum bite force, as does antenna removal. In addition, the performance decrement induced by antennaectomy is abolished by administration of excess OA, and that rescue effect is itself nullified by the simultaneous administration of epinastine. These data show that OA is an important mediator of performance, and thus of performance motivation, in insects, and potentially a promising candidate for the short term manipulation of performance as well.

All crickets used in this study were the second-generation offspring of A. domesticus crickets obtained from a commercial supplier (Fluker’s). Males were raised separately to adulthood on a diet of dried cat food (Purina Cat Complete Chow). All experiments were conducted using crickets of similar ages (i.e. between 12 and 15 days post-eclosion). Following eclosion, we measured maximum bite force of all crickets using standard methods (Losos et al. 2002). Briefly, a Tekscan FlexiForce wireless ETF flexible force circuit was placed between the mandibles of a cricket which would then invariably bite down vigorously on the circuit. We measured bite force 5 times per individual and retained the largest of those five individual measures for analysis, consistent with both standard maximum performance methodology (Losos et al. 2002), and with comparable studies in crickets, including A. domesticus (Condon and Lailvaux 2016; Hall et al. 2010; Lailvaux et al. 2011). Following initial bite force measurement, we allocated crickets randomly to one of five groups: control; epinastine; antennectomized; antennectomized + OA; and antennectomized + OA + epinastine. Crickets then spent two hours inside individual 5x5x5 cm plastic containers with either plain or supplemented excess pureed sweetcorn (as in Bubak et al. 2014; Bubak et al. 2015; Bubak et al. 2013) under one of five conditions depending on their group membership. Control crickets (n = 14) were simply placed in a container with untreated pureed sweetcorn. The sweetcorn of epinastine crickets (n = 15) was supplemented with epinastine at a concentration of 15mg/ml. Antennectomized crickets (n = 18) had both antennae removed with scissors following initial bite force measurement before being placed in a container with untreated pureed sweetcorn. Antennectomized + OA crickets (n = 15) had their antennae removed and were placed in a container containing pureed sweetcorn supplemented with OA at a concentration of 15mg/ml. Antennectomized + OA + epinastine crickets (n = 15) had their antennae removed and were placed in a container with pureed sweetcorn supplemented with both OA and epinastine at the same concentrations as above. Following the two-hour treatments, we measured all crickets for maximum bite force again using exactly the same procedure as before.

All experimental methods complied with the national and institutional ethical guidelines where this work was conducted.

Statistical analysis

We used the nlme R package (Pinheiro et al. 2019) to fit a mixed-model with bite force as a dependent variable; treatment with the levels described above; measurement with the levels pre- and post-treatment as an independent variable; thorax size as a covariate to account for size effects; and cricket identity as a random factor because pre- and post-treatment bite force were measured on the same animals. We also included an interaction between treatment and measure to test our hypotheses which both pertain to significant differences between pre- and post-treatment bite force in certain treatment levels, but not others, and an interaction between size and measure to test for the possibility of size-dependent plasticity in bite force (Lailvaux et al. 2019). P-values associated with individual predictor variables are approximate due to penalty factors applied to random effects during calculation of the likelihood function. Consequently, we assessed the significance of predictors using log-likelihood ratio deletion tests to determine the minimum adequate model, which we then re-fit using restricted estimate maximum likelihood (REML) (Silk et al. 2020). To test for specific differences, we used the emmeans package to extract contrasts comparing pre- and post-treatment bite force within each treatment level (Lenth 2019). We used R v 3.6.0 for all analyses (R Core Team 2019).