Part of the rationale underlying the practice of preregistering research hypotheses is the idea that it will protect researchers against hindsight bias (i.e., against overestimating the foreseeability and plausibility of their findings and constructing possibly flawed hindsight theories around them, which may obstruct further research progress). Strangely, this idea has never been empirically tested. Using the self-reported post hoc expectedness of research outcomes as a proxy for hindsight bias, we conducted an (itself preregistered) survey of posters (N = 351) at two scientific conferences. Poster presenters were asked how much they thought their findings were as expected, and if they had preregistered their study. Unexpectedly, we found that preregistration was unrelated to expectedness (r = -.01). We discuss possible (hindsight) explanations of our null findings, including the idea that there may not be much hindsight bias for research expectations that researchers need to be protected from.

Because people judge easy-to-process (or “fluent”) information positively, we expected members of the public would be more positive about funding scientific research proposals written for a general—compared with a scientific—audience. We asked people to read both kinds of proposals and rate each on several dimensions. People reported greater support for funding the proposals written for a general audience, read those proposals faster, and found them easier to understand than proposals written for a scientific audience. Taken together, these results fit with the idea that making science communication more fluent increases the public’s support for funding research.

Women are underrepresented throughout the science pipeline. Computer science (CS) programs increased women’s participation by focusing introductory courses on applications. This may have unintended consequences if applied work is not valued the same as basic/theoretical work. We surveyed faculty attitudes toward applied researchers, and examined who conducts applied research using publication and grant data. Faculty participants read vignettes presenting hypothetical CS and psychology researchers described as doing basic or applied research. Participants judged applied researchers as less likely than basic researchers to publish in prestigious journals; receive promotions, grants, and major awards; and be brilliant, creative, and technically-skilled. Relative to the percentage of women CS faculty and PhD students, women were more likely to do this undervalued work: they were overrepresented in applied conference publications and application-oriented grant awards and under-represented in theoretical ones. We suggest how academia can better value applied work and the women who do it.