Life history theory is a powerful framework that has been incredibly productive in the empirical study of human behavior. This productivity has brought growing ambiguity about what the theory specifically predicts and under what conditions. Ellis et al.’s two-tiered model provides a much-needed clarification and advancement of our understanding of the intertwining effects of energetic constraints and extrinsic mortality. However, with great theory comes great responsibility, especially the responsibility to spell out conditions under which the model’s predictions can be clearly supported or refuted. We foresee two potential difficulties for clear, falsifiable, empirical tests.

The first challenge concerns how Tier 2 (ambient mortality cues) and Tier 1 (energetic constraints) processes may be sequentially activated by the same ecological event. Ellis et al. interpret fertility increases following major natural disasters as support for the 2-tiered model, noting that these increases occurred despite deteriorating energetic conditions, as evidenced by reductions in childhood growth. This suggests that Tier 2 effects may operate even under energetic hardship. Although Ellis et al. acknowledge that Tier 1 and Tier 2 processes can co-occur, the model provides limited articulation of how their effects may unfold sequentially over time. Specifically, it remains unclear how a single ecological event might initially elicit Tier 2 responses but later produce Tier 1 effects as resource depletion accumulates.

Ecological conditions may simultaneously or sequentially activate both Tier 1 and Tier 2 processes, making predictions highly sensitive to the time window under consideration. To illustrate, in the aftermath of a natural disaster, Tier 2 effects are expected to accelerate reproductive timing, as Ellis et al. describe. However, if the same event also leads to sustained resource depletion, Tier 1 effects should, over time, suppress reproductive rates. For example, over a decade past the disaster occurring, tsunamis have detrimentally impacted the safety of local water sources (Villholth & Neupane, Reference Villholth, Neupane and Mokhtari2011) and earthquakes, such as the 2010 disaster in Haiti, have depressed economic growth (Joseph, Reference Joseph2022). Theoretically, both acceleration and delay in reproduction could be explained within the model, depending on when and how outcomes are measured.

This concern is not purely hypothetical. For instance, the meta-analysis by Lee et al. (Reference Lee, Batyra, Castro and Wilde2023) found that most fertility effects following disasters were negative, though positive effects were observed for physical disasters. Additional research on Hurricane Mitch demonstrated a short-term increase in fertility during the first two years post-disaster, but this effect dissipated four to six years after the event (Davis, Reference Davis2017). Together, these findings suggest that Tier 2 effects may emerge quickly in response to mortality cues but fade as longer-term energetic constraints accumulate. Crucially, without a more explicit articulation of the relative time course of Tier 1 and Tier 2 effects, either pattern in isolation could be interpreted as Tier 2 or Tier 1 after the fact, running a risk of becoming unfalsifiable. Clarifying the trajectory and timing of hypothesized effects would strengthen the model’s predictive precision and guard against concerns about falsifiability.

The second challenge concerns potential curvilinearity of Tier 2 effects. While Ellis et al. propose that Tier 2 cues tend to accelerate reproduction, if Tier 2 effects do not follow a strictly linear pattern, both the delay and acceleration of reproductive strategies could again be accommodated within the model. For example, moderately high levels of pathogen stress are actually associated with greater parental effort (to protect offspring), while very high levels of mortality lead to diminished parental effort (and more broadly faster reproduction), due to the diminished benefits of investment in any individual offspring (e.g., Quinlan, Reference Quinlan2007). Although Quinlan’s findings concern investment behavior rather than reproductive outcomes directly, they suggest that the same Tier 2 cue may yield seemingly opposing outcomes depending on the extremity of the cue. This pattern highlights the possibility that Tier 2 effects may follow a nonlinear trajectory.

Ellis et al. seem to acknowledge nonlinearity when discussing how very high ambient mortality is often accompanied by energetic deprivation and suggest that accelerated reproduction in such cases is constrained by Tier 1 factors (p. 41). This explanation attributes the nonlinear response to the interaction between Tier 1 and Tier 2 effects. Our concern is somewhat different: Tier 2 cues alone may produce nonlinear responses even without constraints posed at Tier 1. If so, clarification is needed about the conditions under which nonlinear Tier 2 responses are expected. One way to better understand potential nonlinearity in Tier 2 effects is to apply a dose–response framework, which examines how the intensity of exposure to a cue affects the nature or strength of the resulting response. This approach would allow researchers to test whether there are threshold points at which exposure to ambient mortality cues initially accelerates reproduction but later reverses direction. Identifying such inflection points would allow the model to generate specific predictions about how reproductive outcomes should vary across different levels of ambient mortality.

To strengthen the two-tiered model’s predictive precision in practice, we recommend clearer articulation of the potential time-dependent dynamics and nonlinear effects. Specifically, the model should clarify when Tier 2 effects accelerate reproduction before Tier 1 constraints slow it, if at all. Likewise, if Tier 2 responses vary by cue intensity, mortality thresholds should be specified in advance. Expanding the model to account for how the same ecological event may sequentially trigger Tier 2 and then Tier 1 processes over time, as well as addressing whether Tier 2 effects follow a nonlinear trajectory, could enhance the falsifiability of the model and guide future empirical work. A theory is only useful to the extent that it can be clearly tested and falsified. We do believe the two-tiered model meets that standard, and addressing the issues above will both strengthen and sharpen it.