jpsp-theory-and-hypotheses - SKILL.md Agent Skill

name: jpsp-theory-and-hypotheses description: Use when building the theoretical argument and deriving testable hypotheses for a Journal of Personality and Social Psychology (JPSP) manuscript, where theoretical innovation is the primary bar and each study in the package must test a specific derivation. Structures the theory; it does not invent constructs or results.

Theory & Hypotheses (jpsp-theory-and-hypotheses)

JPSP's first gate is theoretical innovation: reviewers ask whether the paper's underlying ideas are interesting and generative enough, often independent of how strong the data are. A package of well-run studies with a thin idea will not clear the bar. This skill builds the theory and the hypotheses each study tests; the studies themselves are designed in jpsp-study-design.

When to trigger

Turning a positioned gap into a precise theoretical account
Deriving hypotheses that map onto specific studies in the package
A reviewer questioned the novelty, mechanism, or generativity of the idea

Ways a JPSP paper can be theoretically innovative

(Adapted from JPSP section guidance; verify per section.) A contribution can:

develop a new theory and provide evidence for it;
use an existing theory to explain a new phenomenon;
make novel connections between two theories to address new questions;
use a theory to integrate previously unconnected phenomena;
offer a new mechanism for an established phenomenon;
specify moderators that reconcile conflicting predictions or define when an effect occurs;
introduce and validate a new construct and show it matters;
examine an existing theory or phenomenon in an understudied population.

Building the argument

State the core claim in one sentence, then the mechanism (the why).
Derive hypotheses that distinguish your account from alternatives. Each hypothesis should be one a competing theory would not predict — that is what makes the test diagnostic.
Map hypotheses to studies. H1 → Study 1 (establish), H2 → Study 2 (mechanism/mediation), H3 → Study 3 (moderation/boundary), etc. The package should build cumulatively.
Pre-empt salient alternative explanations (construct validity, alternative mediators, alternative causal models) and design at least one study to rule them out.
State scope and generalization — to which populations/contexts the theory should and should not extend.

Anti-patterns

Hypotheses so vague any result confirms them (unfalsifiable)
A "mechanism" that merely restates the effect in different words
Studies that all test the same hypothesis instead of building cumulatively
HARKing: presenting exploratory findings as if they were predicted (see jpsp-data-analysis)
Ignoring the obvious alternative account a reviewer will raise

Clearing the innovation gate: a worked derivation

Illustrative — invented to model diagnostic hypotheses, not a real theory.

Claim: incidental gratitude broadens construal because it shifts attention toward self-transcendent goals. The diagnostic test is what separates this from a "clean effect" that fails the gate:

H1 (establish, S1). Gratitude raises abstract construal vs. neutral. A pure-affect account predicts the same, so H1 alone is not diagnostic.
H2 (mechanism, S2). Self-transcendence mediates the effect — a prediction a generic positive-affect account does not make. This is the diagnostic hypothesis.
H3 (boundary, S3). The effect attenuates under time pressure, which taxes the goal-shift but not affect — separating mechanism from mood.

Each hypothesis maps to one study, and H2/H3 are built so a competing theory predicts otherwise; that diagnosticity is what an ASC reviewer means by "theoretically generative" — the difference between a JPSP package and a demonstration.

Operating pass for Journal of Personality and Social Psychology

Treat this skill as an executable review pass, not a prose hint. First lock the construct validity, study sequence, power/robustness plan, and boundary conditions; then judge whether the current manuscript answers the venue's real reader: psychology reviewers who need a theoretical construct, validated measurement, and cumulative-study logic.

Do the pass: Return a claim-evidence-risk ledger rather than a prose-only diagnosis; every recommendation must point to a manuscript location or missing artifact.
Return a ledger: give claim / evidence / risk / manuscript location rows, so the next agent can edit rather than rediscover the issue.
Sibling guard: compare against Psychological Science for shorter general-interest findings, Personality and Social Psychology Bulletin for field scope, Journal of Experimental Social Psychology for experiment-centered claims; if a sibling owns the contribution, recommend re-routing before polishing format.
Stop condition: do not give submission-ready advice until the pack's resources/official-source-map.md has been checked for volatile rules and the manuscript has one concrete fix for the largest venue-specific risk.

Output format

【Core claim】one sentence
【Mechanism】the "why"
【Innovation type】new theory / new mechanism / integration / moderator / new construct / …
【Hypotheses → studies】H1→S1, H2→S2 (mechanism), H3→S3 (boundary) …
【Diagnostic vs alternatives】which alternative each test rules out
【Scope / generalization】where the theory applies
【Next】jpsp-study-design

Supplementary resources

../../resources/external_tools.md — SEM / mediation / process tooling for testing derivations
../../resources/official-source-map.md — section innovation criteria and review gate