Neural Erosion: Emulating Controlled Neurodegeneration and Aging in AI Systems
This work addresses the problem of emulating brain function decline for applications in cognitive disorder research, though it is incremental as it extends neurodegeneration modeling from computer vision to text data.
The paper tackled simulating neurodegeneration in AI by applying controlled damage like synapse ablation or noise to Large Language Models (LLMs) during IQ tests, resulting in a progressive decline where LLMs first lose mathematical abilities, then linguistic abilities, and finally question comprehension, with findings aligning with human clinical studies.
Creating controlled methods to simulate neurodegeneration in artificial intelligence (AI) is crucial for applications that emulate brain function decline and cognitive disorders. We use IQ tests performed by Large Language Models (LLMs) and, more specifically, the LLaMA 2 to introduce the concept of ``neural erosion." This deliberate erosion involves ablating synapses or neurons, or adding Gaussian noise during or after training, resulting in a controlled progressive decline in the LLMs' performance. We are able to describe the neurodegeneration in the IQ tests and show that the LLM first loses its mathematical abilities and then its linguistic abilities, while further losing its ability to understand the questions. To the best of our knowledge, this is the first work that models neurodegeneration with text data, compared to other works that operate in the computer vision domain. Finally, we draw similarities between our study and cognitive decline clinical studies involving test subjects. We find that with the application of neurodegenerative methods, LLMs lose abstract thinking abilities, followed by mathematical degradation, and ultimately, a loss in linguistic ability, responding to prompts incoherently. These findings are in accordance with human studies.