2. According to Galton, people incapable of forming images were overrepresented in math and science. I've since heard that this idea has been challenged, but I can't access the study.

The challenging paper: Brewer and Schommer-Aikins (2006)

Abstract:

In 1880, Galton carried out an investigation of imagery in a sample of distinguished men and a sample of nonscientists (adolescent male students). He concluded that scientists were either totally lacking in visual imagery or had “feeble” powers of mental imagery. This finding has been widely accepted in the secondary literature in psychology. A replication of Galton’s study with modern scientists and modern university undergraduates found no scientists totally lacking in visual imagery and very few with feeble visual imagery. Examination of Galton’s published data shows that his own published data do not support his claims about deficient visual imagery in scientists. The modern data for scientists and nonscientists and the 1880 data for scientists and nonscientists are in agreement in showing that all groups report substantial imagery on recollective memory tasks such as Galton’s breakfast questionnaire. We conclude that Galton’s conclusions were an example of theory-laden interpretation of data based on the initial responses from several very salient scientists who reported little or no visual imagery on Galton’s imagery questionnaire.

Conclusions:

It now appears that Galton’s strong claims were incorrect. It is not the case that most scientists show little or no mental imagery. Galton’s own data and our more recent data demonstrate that scientists show strong visual imagery in recollective memory tasks, just as nonscientist undergraduates do. The data do suggest there may be some small differences in vividness of visual imagery between scientists and undergraduates. However, these differences could easily be due to age differences (Galton, 1879, p. 432, suggested that there may be a decline in imagery with age) or to differences in style of reporting internal mental states. It seems to us that future work on these issues should not focus on imagery in recollective memory tasks such as the breakfast questionnaire. It is not obvious that this type of memory plays a special role in the work of scientists. However, we think that there might be interesting differences on various types of spatial reasoning tasks between scientists and nonscientists, and more particularly among different types of scientists (e.g., crystallographers vs. physiologists).

We also think this analysis of the reasons for the discrepancies between Galton’s claims and his data provides interesting insights into the power of top-down factors in the work of scientists. We entertained the hypothesis that the discrepancy was due to deep-seated beliefs about a hierarchy of intellectual abilities. However, we discarded that hypothesis as Galton gave a non-nativist account of his findings and was surprised by his initial finding that a few scientists reported that they had little or no mental imagery.

We conclude that Galton’s top-down interpretation of his findings was not driven by deepseated theoretical beliefs but merely by the occurrence of a few unusual individuals in his pilot sample. If our interpretation is correct, it certainly highlights the powerful role of even relatively routine top-down beliefs in the way that scientists carry out their work (cf. Brewer & Lambert, 2001).