Similarity scores and biases.

These are the similarity scores to the concepts of male, female, posemo, and negemo, as well as gender and valence biases of 11 colour terms and four anchor words. The position of each score in relation to the normative word population (i.e., 100,000 most frequent words) appears under “percentile”. Values in bold are below the 5th or above the 95th percentile, indicating extreme similarities or biases of these words (same data as in Figs 1 and 2).