Connections between language and music have been made for over a century, and their origins are thought to be closely linked, but the selection mechanisms of the first precursors to music and language are still largely unknown, and it is debated whether they would have been either sexually or socially selected for. The present study presents the results of an emotional priming experiment in an attempt to shed light on the selection mechanisms which would have been at work on the evolution of a musical protolanguage. Participants were presented with one of three stories to prime them for sexual, social or neutral feelings and were then asked to rate pieces of music, created by participants of a prior experiment, on a scale for quality; the ratings for the rhythms were compared against cognitive fitness variables of the rhythm creators. A positive correlation was found between ratings of rhythms and the intelligence of the creators for all priming groups (p<0.01), supporting previous findings which suggest that musical signals may be used as a marker of cognitive fitness. A significant difference was also found between the ratings of the social priming group and those of the sexual priming group (p<0.01) and neutral priming group (p<0.05) when correlated with creator processing speed, providing tentative support for the hypothesis that social selection was at least as important as sexual selection in the evolution of a musical protolanguage. These results offer a promising beginning for the formation of arguments in favour of the social selection hypothesis of a musical protolanguage, and demonstrate the need for more research into this theory to be carried out to either disprove or confirm these preliminary findings.