The key idea is that Jelinski–Moranda models the failure rate as proportional to the number of latent faults still present. At any time, the instantaneous failure rate (hazard) is determined by how many latent faults haven’t been found yet. This is expressed as λ(t) = φ (N0 − m(t)), where N0 is the total number of latent faults at the start and φ is the defect-detection rate. As faults are discovered, m(t) increases, leaving fewer latent faults (N0 − m(t)), so the hazard declines over time. This structure places the model in the non-homogeneous Poisson process family but with a specific form for the rate that directly ties to remaining faults. The statement captures this defining relation, making it the best description. The alternative that describes an increasing failure rate conflicts with the decreasing hazard as faults are eliminated. The exponential mean-value form m(t) = a(1 − e^{−bt}) belongs to a different model (Goel–Okumoto). While Jelinski–Moranda can be viewed in the NHPP framework, the crucial distinguishing feature is the λ(t) = φ (N0 − m(t)) form with N0 latent faults and φ as the detection rate.

Persistence in learning is shown when someone sticks with a difficult task, keeps trying, and learns from each attempt until they succeed. Stacking blocks again and again until the tower stays upright demonstrates that steady, repeated effort to improve a skill—continuing to try after failures and adjusting approach until achieving a stable result. This embodies the willingness to endure challenges and refine technique over time, which is at the heart of persistence. The other examples show useful behaviors—exploring different ways to use a tool or repeating practice on a puzzle—but they don’t capture the same clear, sustained commitment to overcoming a specific difficulty with repeated, goal-driven effort. Walking away from a tough task is the opposite of persistence.

This question tests transfer of learning—the ability to take something seen in everyday life and apply it in a new situation. The child watches a police officer demonstrate traffic directing signals and then uses those same signals on the bike track. That shows taking knowledge from a real-world experience and adapting it to a different context, which is exactly what applying everyday experience to a new situation looks like. The other scenarios are less about applying everyday experiences to a novel context. For example, mentioning Nana’s chair shows recognizing similarity to a familiar object but not using that experience to handle a new task. Watching a teacher and then sorting by size demonstrates copying a procedure learned in a setting that’s still familiar, rather than applying everyday knowledge to a new scenario. Sorting crayons by color is basic categorization, not necessarily drawing on everyday experience as a guide in a new situation.

The expression models a parallel arrangement where a series path of two components is in parallel with a third component. A series pair has reliability R1 × R2 because both components must work for that path to succeed. The whole system then works if either that series path works or the individual third component works (or both), which is the essence of parallel redundancy. For parallel paths, the system fails only if every path fails. The series path fails with probability 1 − (R1 × R2), and the third component fails with probability 1 − R3. Assuming independence, the probability that both paths fail is (1 − R1×R2) × (1 − R3). Therefore, the overall reliability is 1 minus that failure probability: R_sys = 1 − (1 − R1×R2)(1 − R3). This aligns exactly with the given formula, confirming the configuration is a series path of R1 and R2 in parallel with R3. If all three were in series, reliability would be R1 × R2 × R3. If all three were in parallel, reliability would be 1 − (1 − R1)(1 − R2)(1 − R3). A single component with no redundancy would just be one R value.