👋 Hi, I’m Melissa and welcome to my biweekly Note, Language Processor. Every other Tuesday, I dig deep into language & behavior, the limits of technologies, and the connection between what people say and do. Once a month, Signal Lab takes over and reports on these topics for startups and early-stage investors.
Why
In the past three years, the breakthrough of LLMs in technology and culture has re-inspired awe for language and a new awareness. We now appreciate complexities we took for granted and realize language contains more than we imagined within it.
But when it comes to language and behavior (i.e., behavioral linguistics), we have hardly scratched the surface.
What
Language Processor exists to:
Spotlight language itself, which we use to persuade, inform, and motivate — and by which we are persuaded, informed, and motivated.
Apply the science of language to real life, and use it to train, challenge, revise, and ultimately enhance intuition and decision. In short, to power your human language processing.
Who
I’m a lifelong word nerd.
But I’m most interested in why people talk the way they do and what those choices reveal.
I started as a researcher then theorist, trying to grasp what I could only sense — an intuitive connection between speech and behavior. This led to years of “convex tinkering” growing from and extending beyond behavioral linguistics. The end result was my own methodology and applied practice.
And so, for the past ten years, I’ve worked in strategic analytics and decision support, in high stakes environments where lives and livelihoods are on the line — expert domains where many competing factors are in play, and the signal to noise ratio is low.
How
There are many ways meaning is encoded and interpreted through language. In my work, I’ve learned how to evaluate language data (e.g., interviews, statements) for person-based risk — durability, credibility, and potential performance — especially where speaker intent cannot be assumed.
Reading people starts with listening to what they say.
When
Upcoming posts will be on language-based predictive models, major decisions and your “gut”, what makes an apology an apology, and linguistic analysis in Private Equity. Occasionally, I’ll cover tools in linguistic analysis you should know about. First up will be micro Language Models.
Once a month, Signal Lab will take over the newsletter and bring language science into signal detection, with special insights for Venture Capital.
Language Processor will be posted biweekly, on Tuesdays.
Subscribe to get full access to the newsletter.
See you in two weeks.