Cogpsych submission

Submission Preflight (cogpsych-submission)

The last check before submitting through Elsevier Editorial Manager. Two things sink Cognitive Psychology submissions most often: a wrong-shape contribution (single effect, no model/theory, or a model that is never compared) and non-reproducible modeling (fits that can't be regenerated, no code deposit). Verify volatile specifics (length, abstract limit, review model, current editor, declarations) on the official guide for authors first (检索于 2026-06；以官网为准).

When to trigger

• "Submitting tomorrow" — last pass before upload
• Unsure what the scope, format, and declaration requirements demand
• Confirming the abstract, exhibits, model comparison, code deposit, and declarations are in order

Process facts (verify volatile items on the official page)

• Owner / publisher: Elsevier. (检索于 2026-06；以官网为准)
• Portal: Elsevier Editorial Manager (linked from the ScienceDirect guide for authors).
• Identity: a leading cognitive-science journal favoring longer, integrative, model-driven articles; multi-experiment programs combined with formal/computational modeling.
• Scope: attention, perception, memory, learning, language, categorization, reasoning, problem-solving, judgment and decision-making; modeling/neuroscientific approaches and substantive reviews welcome.
• Style: APA-style statistical reporting and references; Elsevier formatting via the guide for authors. Confirm the current reference style and any structured-abstract requirement.
• Open science: authors are strongly encouraged to share data, model/analysis code, and materials; complete the Elsevier research-data statement and declaration of competing interest.
• Review model / length / abstract limits: volatile — confirm single- vs. double-anonymized review, any word/length guidance, and the abstract limit on the live page. 待核实.

Preflight checklist

Scope & shape

• Contribution is a theoretical advance carried by a program (multi-experiment and/or a model)
• If a model is central, it is fit and compared to a rival under matched flexibility
• Not a single-effect / atheoretical paper better suited to a short-report venue

Front matter & format

• Abstract present (confirm length/structure on the official page); keywords included
• APA-style statistics (effect sizes + intervals; model-comparison criteria) and references
• Exhibits show observed data + model fit (not bars of means); model-comparison table included
• Manuscript follows Elsevier formatting/section conventions per the guide for authors

Modeling & disclosure

• Model comparison reported (AIC/BIC/BF or cross-validation) with free-parameter counts
• Parameter + model recovery reported
• Mixed/hierarchical structure where the design demands it, with diagnostics
• Confirmatory vs. exploratory clearly separated

Open science & declarations

• Data deposited + DOI + codebook (or justified restriction)
• Model/analysis code deposited; fits regenerate in a fresh session (run log)
• Materials deposited + DOI
• Elsevier research-data statement + competing-interest declaration + funding/contributions
• Preregistration linked where applicable

Worked micro-example — preflight verdict (illustrative)

Running the recognition-memory program through the gate the night before upload:

Shape:      3 experiments + UVSD vs. DPSD comparison → theoretical advance → PASS
Model:      both models fit (matched flexibility); dBIC + BF reported;
            parameter + model recovery included → PASS
Hierarchy:  hierarchical Bayesian; R-hat ~ 1.00, ESS adequate → PASS
Exhibits:   z-ROC figure overlays observed data + both model curves; model-
            comparison table with free-parameter counts → PASS
Open sci:   data + model code + materials each have a DOI; fits regenerate
            in a fresh session (run log attached) → PASS
Declarations: competing interest (none) + research-data statement + funding
            completed in Editorial Manager → PASS
Verdict:    ready to upload to Editorial Manager.

Last-mile failure modes that trigger a return

Failure	Caught by	Quick fix
Model fit but never compared	scope/shape check	fit the rival; add the comparison table
Fits don't regenerate from code	reproducibility check	deposit seeded code + a fresh-session run log
Bars of means, no model overlay	exhibit check	redraw with observed data + model fit
"Data on request" instead of a DOI	open-science check	deposit and mint a persistent identifier
Competing-interest/data statement blank	declaration check	complete the Elsevier declarations before upload

Volatile specifics (length guidance, abstract limit, review model, current editor, exact declarations) change. Re-confirm every item above against the journal's current guide for authors before relying on it (检索于 2026-06；以官网为准).

Anti-patterns

• Submitting a single-effect/atheoretical paper to a model-driven, long-form venue
• A central model that is fit but never compared to a rival
• Modeling that cannot be reproduced from the deposited code
• Exhibits that hide the fit (bars of means) or assert "best fit" with no table
• Missing Elsevier research-data / competing-interest declarations

Output format

【Shape】theoretical advance via a program; model fit + compared? [Y/N]
【Format】abstract + keywords + APA-style reporting + Elsevier formatting? [Y/N]
【Modeling】comparison + recovery + hierarchy + diagnostics? [Y/N]
【Open science】data + model code (reproducible) + materials + DOIs? [Y/N]
【Declarations】competing interest + research-data + funding? [Y/N]
【Next】await decision → cogpsych-rebuttal on revision

Supplementary resources

• templates/checklist.md — the eight-section pre-submission self-check
• ../../resources/external_tools.md — repositories, DOIs, modeling/reproducibility tooling
• ../../resources/official-source-map.md — official Cognitive Psychology / Elsevier URLs behind every fact in this pack