This article presents an experimental study on the naturally biased association between shape and color. For each basic geometric shape studied, participants were asked to indicate the color perceived as most closely related to it, choosing from the Natural Color System Hue Circle. Results show that the choices of color for each shape were not random, that is, participants systematically established an association between shapes and colors when explicitly asked to choose the color that, in their view, without any presupposition, they saw as the most naturally related to a series of given shapes. The strongest relations were found between the triangle and yellows, and the circle and square with reds. By contrast, the parallelogram was connected particularly infrequently with yellows and the pyramid with reds. Correspondence analysis suggested that two main aspects determine these relationships, namely the "warmth" and degree of "natural lightness" of hues.

No abstract

This article presents an experimental study on the natural association between angles and colour. Specifically, for each angle width -formed by two segments joined at their vertex -studied, participants were asked to indicate the colour that they perceived as most closely related to it, choosing from the NCS Hue Circle. The results show that participants systematically established a natural and consistent association between certain angles and colours, when explicitly asked to choose the colour that they perceived as most naturally related to a given angle presented as two lines. The overall results suggest an association between acute angles and warm colours and obtuse angles and cool colours, confirming Kandinsky's hypothesis. In particular, the strongest relations were found between the angle of 22.5°and yellows, the angles of 45°, 90°and 135°and green-yellows, and the angle of 157.5°and red-blues, when angles were presented on a white background.

The study shows a systematic naturally biased association between percepts and concepts. Specifically, it shows that a series of terms pertaining to an abstract semantic field (related to the frame of ethics in social behaviour) has a nonrandom, highly significant, association with colours (hues). This is the first time that consistent associations between abstract terms and colours have been reported in the general population. The main hypothesis, ie that there appear to be 'hues of concepts', was borne out by the results: the abstract terms considered were coloured with blue/green (ie cool) colours as well as their synonyms, while their antonyms were coloured with red/yellow (ie warm) colours. The association provides information about the nature of abstract concepts and their relationship with perception. It also sheds light on the interrelations among words in semantic domains that, to date, have been studied from only a computational viewpoint.

No abstract