---
name: psci-review-process
description: Use when you need to understand how Psychological Science evaluates a manuscript — anonymized peer review, editorial weighting of robustness, transparency, and preregistration quality, desk-reject and decline-without-review patterns, and the Registered Reports route. Use when stress-testing a paper before submission or interpreting a decision letter. Sets expectations and shapes the paper to survive review; it does not contact editors.
---

# Review Process (psci-review-process)

Psychological Science combines high-impact selectivity with strong credibility checks. Reviewers and
editors weigh not only whether the finding is interesting, but whether it is **robust**, **adequately
powered**, and **transparent**. Knowing this lets you pre-empt the common rejection reasons.

## When to trigger

- Before submitting, to stress-test the manuscript
- Deciding whether to use the **Registered Reports** route
- Interpreting a decision letter and setting expectations

## How review works

1. **Anonymized review.** Initial submissions are anonymized; keep author identity out of the
   manuscript and out of repository links (see `psci-submission`).
2. **Editorial triage.** Editors assess impact, breadth, robustness, and fit; the very tight format
   means weak-fit or thin papers may be declined without full review.
3. **External review** assesses theoretical contribution, design and power, analysis and disclosure,
   and the strength of the claim relative to the evidence.
4. **Transparency is graded.** The **Research Transparency Statement** is shared with reviewers, and
   "**limits on transparency will be a factor in editorial decisions**"; **preregistration quality**
   is also weighed.
5. **Decisions.** Reject, revise and resubmit, or accept; expect substantive revision and frequent
   requests for added robustness, disclosure, or analyses.

## Registered Reports route (strongest for confirmatory claims)

- **Stage 1**: theory + design + analysis plan reviewed **before** data; in-principle acceptance
  commits the journal regardless of outcome if you follow the plan. **Stage 2** reports the results.
  This route protects against publication bias and is well suited to confirmatory and replication work
  (and **RR with Existing Data** for prior-collected data).

## Shape the paper to pass

- Make impact and breadth explicit early; show the result is **robust and well-powered**.
- Disclose fully and share data/materials (or justify exemptions) — credibility signals matter here.
- Separate confirmatory from exploratory analyses honestly.
- Fit the format — reviewers notice when a paper fights the word limit.

## Desk-reject and decline-without-review patterns

The tight format and credibility screen mean many manuscripts never reach external review. Confirm
current categories and limits against the journal's submission guidelines, but recognize these shapes:

| Pattern an editor sees | Likely outcome | Pre-empt it by |
|------------------------|----------------|----------------|
| Surprising effect, single small study, no preregistration | declined or RR suggestion | add internal replication; preregister; report power |
| Narrow-paradigm result, no broad-relevance argument | desk reject (fit) | state who outside the subarea inherits the claim |
| "Data available on request," no DOIs | returned for compliance | deposit with persistent IDs before submitting |
| Over the word format, exhibits dumped at the end | returned to author | design to the format; embed exhibits |
| p-values and stars, no effect sizes/CIs | thin-evidence flag | estimation-first reporting |

## Worked micro-example (illustrative triage)

```
Manuscript: two preregistered attention studies (N = 240; N = 300),
            open data + materials with DOIs, effect sizes + CIs.
Editor read: impact (load-bearing premise), breadth (clinical inheritance),
            robustness (internal replication), transparency (graded — strong).
Likely route: external review, probable R&R for added robustness/disclosure.
Counter-case: same finding, one N = 45 study, no prereg, request-only data
            → likely declined without full review.
```

## How reviewers weigh the evidence (calibration anchors)

- A powered internal replication is the single strongest signal you can send; it converts "interesting
  but fragile" into "credible."
- Transparency is graded, not pass/fail — a candid exemption with an access path reads better than
  silent opacity. Preregistration *quality* (specific, dated, followed) matters more than its presence.
- Registered Reports are the venue's structural answer to publication bias; choosing the route after
  data exist defeats its purpose and reviewers will say so.

## Anti-patterns

- A surprising but underpowered, single-study effect
- Weak or absent transparency (counts against the paper)
- Exploratory results dressed as confirmatory
- Expecting acceptance without a robustness/disclosure-heavy R&R
- Choosing a Registered Report after results exist

## Output format

```
【Impact + breadth】clear early? [Y/N]
【Robustness + power】adequate? [Y/N]
【Transparency】data/materials + statement + preregistration strong? [Y/N]
【Confirmatory vs exploratory】honest? [Y/N]
【Route】Research Article vs Registered Report
【Realistic outcome】reject / R&R / accept
【Next】psci-submission (or psci-rebuttal if decided)
```

## Supplementary resources

- [`../../resources/official-source-map.md`](../../resources/official-source-map.md) — review model, transparency weighting, Registered Reports