The color communication game: how categorical understanding of colors can be shown without considering color naming data

There is clear diversity among speakers of a typical language in how colors are named. What is the impact of this diversity on people’s ability to communicate about color? Is there a gap between a person’s general understanding of color terms in their native language and how they use a particular term to denote a particular color sample? Seventy English-speaking dyads and 63 Somali-speaking dyads played the Color Communication Game, where the “sender” in each dyad named 30 color samples as they would in any color-naming study, then the “receiver” chose the sample they thought the sender intended to communicate. English speakers played again, under instructions to intentionally communicate color sample identity. Interpersonal Mutual Information (MI) calculated from color naming data was variable and below optimal, and English-speaking dyads’ MI did not improve with experience. Although Somali-speaking senders provided fewer color terms, both groups revealed a superior understanding of color terms because they showed better exactly correct selection performance as receivers than was predicted by simulation from their dyads’ color-naming data. Direct comparison of senders’ samples and receivers’ choices revealed a categorical understanding of colors without considering color naming data. This study highlights limitations on information-theoretic analyses of color naming data.

Methods
In the color communication game, one player is the "sender" and another player is the "receiver". The sender selects a sample from his source set of samples and sends a message (a color term), which is heard by the receiver. The receiver then selects from her destination set of color samples the one she thinks the sender intended to communicate. Neither player can see the other’s samples. The message improves the receiver’s likelihood of selecting the correct sample, but it does not guarantee success. English-speaking and Somali-speaking participants played the game. The 30 colored stimuli were presented on calibrated iPads and the data were recorded on the iPad. The data were tabulated in Excel files (here, CSV files). They were analyzed using Mathematica software. The routine for calculating Mutual Information from the color terms provided by the sender and the receiver is attached here.