A Large Language Model (LLM) is an artificial intelligence system designed to understand and generate human-like text based on vast amounts of data. It uses deep learning techniques, particularly neural networks with many parameters, to perform tasks such as translation, summarization, and conversation.

Since November 30, 2022, OpenAI has already taken the world by storm, with its era-defining ChatGPT (an LLM). The platform received approximately 152 million visitors in it's first month, and inspired many companies joining the AI race. It dramatically changed the way human interacting with the computer.

In order to study LLMs, we need to have a basic conceptual understanding of it. Please watch the Youtube video: Intro to Large Language Models [59:47] by Dr. Andrej Karpathy, an OpenAI co-founder and researcher. The video provides a comprehensive introduction for LLMs from structure, training, future outlooks to security issues.  

Psychology is the scientific study of the mind and behavior, exploring how people think, feel, and act both individually and in groups. It aims to understand mental processes and behavior through research and applies this knowledge to improve individual and societal well-being. 

Large language models (LLMs) are machine learning models that can comprehend and generate human language text. Since they were trained with huge amount of human-generated text on the internet, they can be viewed as the compressed version of human knowledge/intelligence. 

The modern LLMs have reached the point of emulating humans: holding conversations, providing advice, writing poems, and proving theorems. The most recent LLM such as ChatGPT-4 has mostly passed the rigorous Turing Test. They brought many new opportunities and applications to psychology.

There are two interesting areas where psychology and LLMs intersect. (1) The LLM can potentially play the role of a quality conversational agent or a research assistant. It can assist psychologists to conduct assessments and analyze qualitative data, such as interview transcripts. (2) LLMs enables new explorations of the mechanisms underlying intelligence and reasoning of human minds.  

Here are a few academic papers exploring the relationships between psychology and LLMs:

• Probing the psychology of AI models (2023)

• A Turing test of whether AI chatbots are behaviorally similar to humans (2024)

• Large language models could change the future of behavioral healthcare: a proposal for responsible development and evaluation (2024)