英和和英

名詞

adversarial collaboration (usually uncountable, plural adversarial collaborations)

1. (sciences, design of experiments, idiomatic, uncountable) Any of a class of study designs in which scientists with conflicting theories work together with others who are neutral or agnostic about those theories, and this team runs an experiment to test the theories empirically and thus find out whether one of the theories is contradicted by the evidence found.

   Hypernyms: collaboration, cooperation

   • 2025 April 30, Carl Zimmer, “Two Theories of Consciousness Faced Off. The Ref Took a Beating. What makes humans conscious? Scientists disagree, strongly, as one group of peacemakers discovered the hard way”, in New York Times‎:

     Dr. Melloni and a group of like-minded scientists began drawing up plans for their study in 2018. They wanted to try an approach known as adversarial collaboration, in which scientists with opposing theories join forces with neutral researchers. The team chose two theories to test. One, called Global Neuronal Workspace Theory, was developed in the early 2000s by Stanislas Dehaene, a cognitive neuroscientist with the Collège de France in Paris, and his colleagues. […] The other theory, developed by Giulio Tononi of the University of Wisconsin and his colleagues, goes by the name Integrated Information Theory.

2. (countable) An instance of such a study.

   Hypernyms: collaboration, cooperation; study, experiment

   • 2022, Cory J. Clark, Nathan Honeycutt, and Lee Jussim, “Chapter 3: Replicability and the Psychology of Science”, in William O'Donohue, Akihiko Masuda, Scott Lilienfeld, editors, Avoiding Questionable Research Practices in Applied Psychology‎, Springer Nature, →ISBN, page 65:

     Although adversarial collaborations might feel like an unnecessary constraint in the short term, it will likely improve research in the long run (Ellemers et al., 2020). If your hypothesis is correct, it likely will win out in an adversarial collaboration. If it is incorrect, likely it will eventually be falsified regardless of whether you discover this on your own in an adversarial collaboration or whether other scholars discover this in failed replications or failed conceptual replications. Delaying the inevitable by refusing to participate in adversarial collaborations only risks wasting more time and money and lowering the ratio of science that will withstand the test of time.