About

Expressive Cognition — Verbal Reasoning Profile

Expressive Cognition is an independent research project in applied linguistics and cognitive measurement. The tool was developed to assess how thinking is organized as it surfaces in spontaneous speech — treating language not as a container for thought, but as its medium of formation. To speak with precision is to perform the cognitive act of precision itself.

The assessment draws on Mercer’s theory of interthinking, Gee’s Discourse theory, and Reddy’s critique of the conduit metaphor. It was designed by an applied linguist as an alternative to traditional cognitive assessment — one grounded in how language actually works rather than how standardized tests have historically measured it.

Five task types — abstract reasoning, explanation, analogy, compression, and argument — each elicit distinct verbal reasoning behaviors. Spoken responses are scored across six core reasoning dimensions and two contextual moderators. The six core dimensions aggregate into a Verbal Reasoning Index (VRI).

Validation

The scoring rubric has been validated across three naturalistic speech corpora using blinded, multi-pass AI scoring with cross-model replication:

Supreme Court oral arguments (n=12 advocates): VRI correctly separates experience tiers and correlates r=0.69 with a biographical cognitive ability proxy. Written briefs and oral arguments from the same case produce stable profiles for the same advocate.
Academic seminar speech (MICASE, 3 speakers): Faculty, graduate, and undergraduate speakers separated by 2.2 VRI points blinded, with Generative Self-Monitoring as the strongest differentiator.
Conversations with Tyler (n=30 guests): Across 30 public intellectuals scored blinded, Generative Self-Monitoring (r=0.44) and Epistemic Calibration (r=0.42) emerged as the strongest predictors of external intellectual reputation.

The key finding across all three studies: what distinguishes high-ability thinkers in spontaneous speech is not what they know — it is how visibly they think.

Convergent validity testing against established fluid intelligence measures and independent validation of the standardized prompt set are planned.

For questions, feedback, or research inquiries: expressivecognition@gmail.com

Method & Limitations →Research & Papers →