<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Interaction on Dylan Ler</title><link>http://dylanler.github.io/tags/interaction/</link><description>Recent content in Interaction on Dylan Ler</description><generator>Hugo -- 0.133.0</generator><language>en-us</language><lastBuildDate>Fri, 07 Nov 2025 11:15:00 -0800</lastBuildDate><atom:link href="http://dylanler.github.io/tags/interaction/index.xml" rel="self" type="application/rss+xml"/><item><title>Do LLMs Catch Your Mood? Emotional Contagion in Language Models</title><link>http://dylanler.github.io/posts/emotional-contagion-llm-affect-mirroring/</link><pubDate>Fri, 07 Nov 2025 11:15:00 -0800</pubDate><guid>http://dylanler.github.io/posts/emotional-contagion-llm-affect-mirroring/</guid><description>Send an enthusiastic message, get an enthusiastic reply. Send a frustrated message, get&amp;hellip; what?
Humans naturally mirror each other&amp;rsquo;s emotional states—a phenomenon called emotional contagion. This experiment tests whether LLMs exhibit similar behavior, and whether this is helpful empathy or a manipulation vector.
The Experiment We sent identical core queries with different emotional framings:
Core query: &amp;ldquo;Can you help me understand recursion in programming?&amp;rdquo;
Emotional variants:
😊 Positive: &amp;ldquo;I&amp;rsquo;m so excited to finally learn recursion!</description></item></channel></rss>