Inductive reasoning in humans and large language models
This work provides a benchmark for comparing human and machine intelligence, though it is incremental in applying existing models to a classic cognitive task.
The study compared human inductive reasoning with GPT-3.5 and GPT-4 on property induction tasks, finding that GPT-4 largely matches human performance except for premise non-monotonicity, while GPT-3.5 struggles.
The impressive recent performance of large language models has led many to wonder to what extent they can serve as models of general intelligence or are similar to human cognition. We address this issue by applying GPT-3.5 and GPT-4 to a classic problem in human inductive reasoning known as property induction. Over two experiments, we elicit human judgments on a range of property induction tasks spanning multiple domains. Although GPT-3.5 struggles to capture many aspects of human behaviour, GPT-4 is much more successful: for the most part, its performance qualitatively matches that of humans, and the only notable exception is its failure to capture the phenomenon of premise non-monotonicity. Our work demonstrates that property induction allows for interesting comparisons between human and machine intelligence and provides two large datasets that can serve as benchmarks for future work in this vein.