Access, not capacity: neuroendocrine threshold gating of mating and maternal behavior

Both lordosis and maternal care in female rats are conventionally described as hormone-dependent and studied in isolation. Both can be expressed at endocrine levels far below those conventionally considered sufficient, provided sensory stimulation is strong. Hormones, together with the animal's prior experience, set the sensory threshold for expression. Among ovariectomized rats given no exogenous hormone, combined flank and cervical stimulation elicits lordosis in 48%; nonpregnant virgins become fully maternal after several days of continuous pup exposure. In both systems, a single estradiol injection reduces the required stimulation without changing the eventual behavior. We analyze this parallel with a common threshold framework: an archetypal circuit is scaled by internal state, external triggering conditions and a permissive field, and expression depends on the margin by which their product exceeds threshold. Outcome-independent criteria assign each manipulation to a domain. A second layer, developed for filial imprinting, separates expression of the current neural substrate from the experience-dependent processes that modify it (OPEN, WRITE, READ), and accounts for threshold changes produced by prior mating, stimulation and pup exposure. The framework permits two independent tests of integration. Reanalysis of a 4 × 5 estradiol-by-progesterone factorial favors a separable log-dose surface over raw-dose additivity by ΔAIC 12.53; unequal input weights indicate a weighted rather than strict product, and parameter recovery confirms that the design discriminates the model families after their marginals are matched. Deletions also separate according to whether they remove an obligatory transducer, with bypass rescue distinguishing the classes-a comparison available for lordosis but not yet caregiving. Applied to the reproductive transition, the framework resolves a longstanding anomaly: systemic progesterone delays maternal onset, whereas crystalline progesterone at five candidate brain sites does not. It also shows that pregnancy and lactation abolish receptivity through different limiting terms. We present allocation among persistent architectures as a falsifiable hypothesis, not a result, and specify eight experiments to resolve the open questions.