Level 5 · Posteriori · L₀(t+1)
θ′ — The Updated Prior
The evidence brief is not the output of the loop. It is the input to the next iteration — θ′, the updated prior that makes the next search better, the next appraisal sharper, the next recommendation more honest.
Four levels built the machine and mapped its error structure. Level 5 is about what the machine produces — not as a terminal output but as a prior update. Each brief you write changes what you know before the next brief begins. Each surveillance check shifts the loss landscape. Each institutional integration expands the team's collective θ. The loop does not terminate. It converges.
A well-produced WHO India evidence brief does not just inform one decision. It updates the team's prior across multiple dimensions simultaneously. Here is what changes after a brief is produced and circulated:
| What is updated | How the brief changes it | How to capture the update |
|---|---|---|
| The evidence baseline | The search protocol log becomes the documented baseline for the next search. What was retrieved, what was missing, and what the evidence supported at this moment in time is now on record. | File the protocol log with a date stamp. The next brief on the same topic starts with this log as its prior, not from scratch. |
| The team's PECO-F vocabulary | Each brief refines the population, exposure, outcome, and equity terms that produce relevant results for WHO India topics. A well-calibrated PECO-F from Brief N becomes the starting framing for Brief N+1. | Maintain a shared PECO-F library. After each brief, add the most productive framing to the library for reuse. |
| The known ε map | Every structured absence named in a brief is a documented gap. The next brief starts knowing which populations lack evidence — it does not rediscover this from scratch. | |
| The institutional prior | A brief that reaches MoHFW and is acted on changes what the Ministry knows and expects from future WHO India briefs. The institutional relationship is itself a parameter — and briefs update it. | Note any Ministry response to the brief. What questions did they ask? What did they push back on? These are gradient signals for the next iteration. |
| The anchor constraint definition | Anchor populations identified in one brief may be more precisely defined in the next, as evidence accumulates. The constraint sharpens over iterations. | Update the anchor population definition in the team's shared SOP after each brief. A more precise constraint is a stronger institutional commitment. |
The most durable practice in this session is the simplest: after every brief, write a one-page prior update document. It takes 20 minutes and transforms a terminal output into a feed-forward for the next iteration. The update document has five fields:
Describe the brief you just completed and any response it received. The tool will produce a structured prior update document — the five fields that make this brief the starting point for the next one.
θ′ is not the end — it is the beginning of the next iteration. A brief that does not produce a prior update document is a terminal output. A brief that produces one is a feed-forward. Five things a brief updates: the evidence baseline, the PECO-F vocabulary, the known ε map, the institutional prior, and the anchor constraint definition. Twenty minutes after circulation, the prior update document exists. The next brief starts there, not from zero. Session 2 addresses how to make this update durable — surviving staff rotation and institutional memory loss.