Animals frequently make adaptive decisions about what to prioritize when they are faced with multiple, conflicting demands at the same time. However, the proximate mechanisms of decision-making in the face of competing demands are not well understood. We explored this question using brain transcriptomics in a classic model system: threespined sticklebacks, where males face conflict between courtship and territorial defense. The outcome of decisions made by breeding males over whether to court mates or defend against territory intrusion are profoundly important to fitness. We characterized the behavior and brain gene expression profiles of males confronted by a tradeoff between courtship and territorial defense by comparing them to males not confronted by this tradeoff. When faced with the tradeoff, males behaviorally prioritized defense over courtship, and this decision was reflected in their brain gene expression profiles. A distinct set of genes and biological processes was recruited in the brain when males faced a tradeoff and these responses were largely non-overlapping across two brain regions. Combined, these results raise new questions about the interplay between the neural and molecular mechanisms involved in decision-making.

Laboratory populations and housing

All experiments were conducted during the summer breeding season (June – August) of 2020 using lab-reared individuals at the University of Illinois Urbana Champaign. Parents of the F1 individuals were wild caught from a population of the “white” ecotype (Blouw and Hagen 1990; Samuk 2016) of threespined stickleback (Gasterosteus aculeatus) from Canal Lake, Nova Scotia, CA (44°29'54.0"N 63°54'09.1"W). Laboratory conditions were maintained at 20°C and a 16:8 light / dark photoperiod to stimulate male nesting behavior. All individuals were fed ad libitum once a day with a mixture of brine shrimp (Artemia sp.), mysis shrimp (Mysis sp.), bloodworms (Chironomus sp.), and Cyclop-Eeze. Prior to experimental trials, males and females were housed socially (5-20/tank, 53L x 33W x 24H cm), but focal males were moved to individual tanks at least 24 hours prior to experimental trials. N=17 focal males were used in this experiment.

Experimental tank set-up

Each experimental tank (53L x 33W x 24H cm) was filled with a layer of gravel on the bottom and divided into three sections. The middle section contained a clump of algae and a plastic tray filled with sand to provide nesting material, and the outer two sections were used to present stimuli during behavioral trials. To ensure construction of the nest in the middle section of the tank, the focal male was first confined to the middle section using dividers. After the nest was constructed, the dividers were removed so trials could be conducted. 

Overview of behavioral trials

Behavioral trials were conducted in August 2020 to investigate the behavioral mechanisms underlying how males managed the competing demands of courtship and territory defense. Behavioral trials consisted of four different treatments: a courtship opportunity, a territorial challenge, simultaneous courtship opportunity and territorial challenge (hereafter referred to as the “tradeoff” treatment), and a control. At the start of each trial, two flasks were introduced into the focal male’s tank. In the courtship opportunity treatment, a gravid female was confined to one of the flasks (the other flask was empty), in the territorial challenge treatment, a rival male (a male displaying breeding coloration) was confined to one of the flasks, and in the tradeoff treatment, a gravid female was confined to one of the flasks and a rival male was confined to the other flask. Rival males were size matched to focal males as much as possible, which gave an average ratio of intruder standard length to focal standard length of 0.97mm (+/- 0.02 standard deviation). In the control treatment, males were presented with two empty flasks. Each focal male was exposed to each of the four treatments (courtship opportunity, territorial challenge, tradeoff, and control) twice in a randomized block design, for a total of eight trials per focal male. Therefore, each focal male was exposed to each of the four treatments in a random order once, followed by each of the four treatments in a random order a second time. Each male was exposed to only one trial per day. This study design is powerful, as it allowed us to investigate plasticity and repeatability in courtship and aggressive behaviors while controlling for order effects.

The focal male was acclimated to the presence of the observer for five minutes before the start of each trial. For each trial, the side of the tank where the flask containing a stimulus was placed was determined by coin toss (excepting the control treatment, which consisted of an empty flask on each side). The two flasks were then gently placed on their respective sides of the tank and male behavior was scored for five minutes using JWatcher (Blumstein and Daniel 2007). The number of zigzag displays (a courtship display) and the number of bites performed by the male were recorded. Males occasionally performed zigzags toward the territorial challenge or empty flask and bites toward the courtship opportunity or empty flask, however it is not possible to distinguish between mistakes and intentional behavior, therefore the number of zigzags and the number of bites performed include all instances regardless of where they were directed.

Statistical analyses

All analyses were conducted using R version 4.2.2 (R core development team, 2022). To investigate how individuals behaviorally responded to the tradeoff between courtship and territory defense, generalized linear models were fit using the ‘lme4’ package and p-values were calculated using a Satterthwaite approximation with the ‘lmerTest’ package. Two models were fit with 1) the number of zigzag displays when a female was present and 2) the number of bites when a territorial intruder was present, respectively, with treatment as a fixed effect. In the model for zigzags, treatment consisted of the courtship opportunity and the tradeoff treatments, since the territorial challenge and control treatments did not contain a female stimulus. Similarly, in the model for bites, treatment consisted of the territorial challenge and the tradeoff treatments. Individual ID was included as a random effect to control for repeated measures of the same individuals. Zigzags and bites are count variables, therefore models were fit with poisson error distributions and an observation level random effect was included to control for overdispersion (Harrison 2014). To investigate order effects, we fit trial set (the first or second time the set of the four treatments was presented), trial order (whether the stimulus was presented first, second, third, or fourth in the sequence), and the interaction between trial set and trial order as additional fixed effects in the models. We then performed likelihood ratio tests to determine whether each factor and their interaction significantly improved the fit of the model. Next, we estimated adjusted repeatability of zigzags and bites as the proportion of total variance accounted for by differences among individuals, while accounting for treatment effects, using the ‘rptR’ package (Nakagawa and Schielzeth 2010). We fit two models with zigzags in the courtship opportunity and tradeoff treatments and bites in the territorial challenge and tradeoff treatments as the response variables, with individual ID as a random effect. We fit models with poisson error distributions and ‘rptR’ fits an additive overdispersion correction.