Cognitive-Based Design Principles in Experimental Economics: How to Devise More Effective Instructions, Software, and Comprehension Tests

Abstract

Economic laboratory experiments are often used for testing real-world economic mechanisms, such as multi-item auctions and multi-contract cost sharing. Ensuring that participants understand these increasingly complex tasks is crucial in order to maintain control, the reproducibility, and the internal validity of the experiments. Traditionally, experimental economists communicate tasks through written instructions. Learning from instructions by reading, understanding, and memorizing involves advanced cognitive processes. However, economists have as yet not fully integrated the findings of cognitive research into their instruction and experiment design. A large body of previous work has shown that cognitive-based designs substantially increase the effectiveness of learning. This paper seeks to leverage the findings of cognitive research for the design of economic laboratory experiments. In an interdisciplinary approach, it reviews and transfers results between the two fields. It discusses the practical limitations of traditional methods of experimental economics; provides an overview of the related cognitive research; and presents concrete guidelines for devising cognitive-based instructions, software, and comprehension tests for reproducible economic laboratory experiments.