Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

Can Induced Emotion Bias LLM Behaviors in Sequential Decision Making?

Source

arXiv

arxiv.org/abs/2607.12631v1

Read the source

Source

Plain-Language Brief

Researchers investigated if emotions induced in large language models (LLMs) could influence their sequential decision-making using the Iowa Gambling Task (IGT). They found that, unlike humans, induced emotion did not significantly bias the decision dynamics of LLM agents on average. However, anger specifically made LLM agents less sensitive to penalties for bad decisions and decreased exploration in early stages of the game.

What the Document Says

  • The study focuses on understanding factors influencing decision-making in Large Language Models (LLMs) in high-stakes domains.
  • Traditionally, LLMs are trained to perceive and resonate with users' emotions, but it is unknown if induced emotion can affect their sequential decision-making.
  • The researchers use the Iowa Gambling Task (IGT), a classic method for studying human decision-making under uncertainty, combined with an imagination-based emotion induction procedure.
  • They confirm that LLMs can sense emotions from context and learn from sequential interactions at a human-like pace.
  • With the validated setup, they find that induced emotion does not significantly bias the decision dynamics of LLM agents on average.
  • However, anger specifically makes LLM agents less sensitive to penalties for bad decisions and decreases exploration in early stages of the game.

Key Concepts Mentioned

  • Large Language Models (LLMs)
  • Iowa Gambling Task (IGT)
  • emotion induction
  • sequential decision-making

Claims Made by the Document

  • LLMs can sense strong, distinguishable emotions from context.
  • LLM agents can learn from sequential interactions in a human-like pace.
  • Induced emotion does not significantly bias the decision dynamics of LLM agents on average.

Evidence Used

  • Validation through confirming that LLMs can sense emotions and learn from sequential interactions.
  • Findings that induced emotion does not significantly bias the decision dynamics of LLM agents on average, but anger specifically has an impact.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

As Large Language Models (LLMs) are increasingly deployed as autonomous agents in high-stakes domains, understanding contextual factors that may modulate their decision-making becomes critical. While LLMs are trained to perceive and resonate with users' emotions, it remains unclear whether induced emotion can influence their sequential decision-making. We investigate this question using the Iowa Gambling Task (IGT), a classic psychological paradigm for studying decision-making under uncertainty, combined with an imagination-based emotion induction procedure. We first validate the feasibility of this paradigm by confirming that LLMs can sense strong, distinguishable emotions from context and that LLM agents can learn from sequential interactions in a human-like pace. With the validated setup, we find that, different from humans, induced emotion does not significantly bias the decision dynamics of LLM agents on average. However, the effects of anger are conditioned: inducing anger makes LLM agents less sensitive to penalties for bad decisions, and in early stages of the game, anger can lower exploration, locking decisions into a few choices early. These findings reveal the subtle yet distinct effects of induced emotion on LLM decision-making compared to human behavior, and provide a tool for future research on affective modulation of LLM agents.