Behavior study experiments are an important part of society modeling and understanding human interactions. In practice, many behavioral experiments encounter challenges related to internal and external validity, reproducibility, and social bias due to the complexity of social interactions and cooperation in human user studies. Recent advances in Large Language Models (LLMs) have provided researchers with a new promising tool for the simulation of human behavior. However, existing LLM-based simulations operate under the unproven hypothesis that LLM agents behave similarly to humans as well as ignore a crucial factor in human decision-making: emotions. In this paper, we introduce a novel methodology and the framework to study both, the decision-making of LLMs and their alignment with human behavior under emotional states. Experiments with GPT-3.5 and GPT-4 on four games from two different classes of behavioral game theory showed that emotions profoundly impact the performance of LLMs, leading to the development of more optimal strategies. While there is a strong alignment between the behavioral responses of GPT-3.5 and human participants, particularly evident in bargaining games, GPT-4 exhibits consistent behavior, ignoring induced emotions for rationality decisions. Surprisingly, emotional prompting, particularly with `anger' emotion, can disrupt the "superhuman" alignment of GPT-4, resembling human emotional responses.

Much of our current understanding about Artificial Intelligence is informed by what we see in popular culture. Due to ignorance and popular cultural portrayals of A.I. and machine learning, many people fear machines that are capable of computing complex tasks. There exists the belief that artificial intelligence could operate in a way that is counter to humanity's best interests, conjuring images of The Terminator and other similar films. As we explore the boundaries of the technology with consumer-facing technology like self-driving cars, and machine-learning algorithms such as recommendation engines, the general public is learning more and more about how these things work. But will we get to a point where people accept AI? A.I. refers to software that models its programming on human behaviour, mimicking rational and logical decision making processes.

Sad, mad, glad, confused about the presidential race so far? Well, if you are really lost on who to vote for, IBM may have a solution, depending on your mood. Watson Elections, one of the surprisingly few political applications for IBM's supercomputer, debuted on stage this morning at TechCrunch's Disrupt Hackathon in Brooklyn, New York to help us decided the best candidate fitting our feelings. If you are angry and disgusted right now, you may want to go for Donald Trump. A little less angry and just sadder fits a vote for Bernie Sanders.