Table 1 Predictions for potential moderators for mean and variance terminal investment effect (note that the inclusion of a particular moderator in our analysis is subject to whether there are sufficient sample sizes in our final dataset; see “Methods” section for details)
Moderator | Predicted mean effect | Predicted variance effect | Rationale for predictions |
|---|---|---|---|
Average lifespan of species | Shorter-living species > longer-living species | Longer-living species > shorter-living species | Mean prediction: Longer-living species are likely to have greater future reproductive prospects compared to shorter-living species. Therefore, they might exhibit a lower terminal investment effect Variance prediction: Longer-living species have greater phenotypic plasticity in general [1]. Therefore, we expect the variance effect to be greater in longer-living species |
Age class | Older > Younger | Mixed age > Young or old | Mean prediction: Compared to younger individuals of the same species, a given survival threat is more likely to tip older individuals past their terminal investment threshold and cause them to increase reproductive investments [13]. There is some evidence supporting this prediction (e.g. [24]) Variance prediction: We expect mixed age samples to contain individuals in different life-history stages. Therefore, such samples will show a greater variance in reproductive investment responses compared to more homogenous samples. We do not have any specific predictions for young versus old individuals because there are many potential situations that can influence differences in variance between the two groups. Young categories may include short- and long-lived individuals, whereas old classes only long-lived ones. In such cases, the variance will be greater in the young individuals. Alternatively, young individuals may all lean towards investing less in current reproduction in favour of future reproductive prospects. Here, we might see a smaller variance in younger individuals |
Whether focal sex provides extended parental care | No extended parental care > Extended parental care | No specific predictions | Mean prediction: Individuals with extended parental care might be less inclined to invest in reproduction during a simulated infection because of the possibility of transferring potentially harmful infections to the offspring [17] |
Control procedures used | Control procedures used that do not invoke an immune response > control procedures used that might invoke an immune response (e.g. via wound healing through sham surgical implants or injections) | No specific predictions | Mean prediction: Some control procedures may inadvertently invoke an immune response via wound healing [16] and consequently cause the control individuals to upregulate their reproductive investments. One example is sham implants, where the control group is subject to all the surgical procedures of an implant except that the implant does not contain the active test ingredient. The lacerations sustained during the sham procedure may invoke an immune response via wound healing. We expect the terminal investment effect size to be smaller in such studies |
Experimental setting (Laboratory versus wild experiments) | No specific prediction | No specific predictions | Mean prediction: We might expect the terminal investment effect to be stronger in laboratory studies, where confounding variables can be better controlled. Alternatively, wild individuals might be more likely to show a terminal investment response due to their higher mortality rates (i.e. lower terminal investment threshold) [25] Variance prediction: Populations brought into the laboratory are likely to face more homogenous environmental conditions during the experiment. Depending on whether the laboratory environment was suited to the needs of all individuals in the sample, we may observe reduced or increased variation in the response to the immune challenge |
Source of the animals (wild species, wild-caught and bred in the lab for several generations, and cultured commercial species) | No specific predictions | No specific directional predictions | Variance prediction: Individuals from the wild might be more varied due to environmental differences. Therefore, they might show more variance in their responses to the immune challenge. Alternatively, wild populations might already have hit a ceiling in their phenotypic variability. In this case, we might expect the more homogenous populations (e.g. those that have been bred/cultured for several generations) to respond with increased variance |
Reproductive investment categories (pre-mating sexually selected physical traits, mating/courtship behaviours and effort, parental care provisioning, reproductive output, later-life offspring traits, and post-copulatory traits) | No specific predictions | No specific predictions | |
Type of immune challenge (non-pathogenic foreign bodies versus substrates of pathogenic origins) | No specific predictions | No specific predictions | |
Taxonomic group | No specific predictions | No specific predictions |