Receptiviti Labs
Independent psychological measurement for AI systems.
Receptiviti was founded in 2015 to measure psychological state from language. The framework underneath it, LIWC, was created by co-founder Dr. James W. Pennebaker and validated across 34,000 peer-reviewed studies.
AI introduced a new requirement in human-facing applications: representing the user’s state as an observable variable outside the model, rather than leaving it implicit in the model’s inference. Made explicit, it can be inspected, reused, and supplied back to AI systems as structured context.
Receptiviti Labs develops the research, methods, and infrastructure for that representation.
34,000+
Peer-reviewed research citations
200+
Validated psychological dimensions
30+
Years of foundational research
THE PROBLEM
Every AI system forms an internal representation of the person it’s talking to.
Whether someone is confused, overwhelmed, distressed, confident, or following the reasoning. That representation influences every response, but it remains implicit inside the model. Three consequences follow.
INSPECTION
Nobody can see it.
Evaluators can inspect prompts and responses, but not the representation that produced them. If the system misunderstood the user, there is no observable variable to verify, challenge, or compare.
RESPONSE
The model acts on it anyway.
Every response is conditioned on that representation. When it’s wrong, the system adapts to the wrong person, changing explanations, confidence, reassurance, pacing, or safety behavior based on an estimate that no one has inspected.
PERSISTENCE
The representation disappears.
Once the response is generated, the model’s representation is gone. It can’t be reused, compared across systems, tracked over time, or supplied to another model.
WHAT MEASUREMENT CREATES
Measured once. Used three ways.
INSPECTION
Independent evidence
Evaluation gains an observable variable for the person on the other side of the interaction, scored alongside the model-quality metrics the stack already collects. The same language always produces the same scores, and every measure is grounded in published psycholinguistic research.
RESPONSE
Structured context
Instead of asking the model to infer psychological state from scratch every turn, measured state becomes structured context that can guide reasoning, pacing, safety behavior, escalation, or handoff.
PERSISTENCE
Persistent state
Measured state becomes a stable reference across conversations, model versions, and time. That enables longitudinal measurement of learning, dependence, wellbeing, rapport, and other trajectories that cannot be observed from isolated responses.
+4%
improvement in educational effectiveness. GPT Study Mode supplied with a 12-signal psychological vector as context. 25 evaluations, five blinded raters, no retraining. Largest gains in reasoning and scaffolding (+6.3%) and cognitive-load management (+5.9%).
The model already had the capability. What it lacked was reliable visibility into the person it was responding to.
INFRASTRUCTURE
Language in. Measured state out.
Dimensions
Deterministic
Model-independent
Latency
Deployment
Integration
Reusable
200+ validated psychological measures
The same input returns the same score
Unchanged by model version or phrasing
Under 65 ms, fast enough to run per turn
Hosted API or on-prem container
Appended directly to existing logs and traces
One representation, consumed by evaluation, inference, and governance
WHY IT WORKS
Stable representations require stable measurement.
A representation that changes every time the underlying model changes cannot serve as an independent reference.
Receptiviti computes psychological state from a validated psycholinguistic framework developed over three decades of research, producing reproducible measurements that remain comparable across models, deployments, and time.