The blue area requires multiple colour names in Italian
Galina Paramei and Cristina Stara
The blue area of colour space arguably requires more than one basic colour term (CT) in Italian [Paggetti et al, 2011, Attention, Perception & Psychophysics, 73, 491-503]. This proposition was addressed in a colour mapping task employing Munsell 7.5 BG–5 PB charts to explore the frequency and consistency of CTs used by Italian speakers compared to English speakers. Participants were Italian monolinguals (N=13; Sassari), English monolinguals (N=13; Liverpool) and Italian-English bilinguals (N=13; Liverpool); the latter completed the task in both languages. Munsell chips were labelled using the unconstrained colour naming method. Participants then indicated the best example (focal colour) of frequent monolexemic CTs (e.g. turquoise, blue for English; turchese, azzurro for Italian). For these, ‘3D Munsell maps’ were constructed. Italian speakers were found to require at least three CTs, with the most frequent and consistent use of celeste ‘light blue’, azzurro ‘medium blue’ and blu ‘dark blue’. Compared to the English focal blue, the Italian focal blu appeared to be darker. Notably, in bilinguals it was shifted towards the English focal blue, with the extent of the shift related to proficiency in English and duration of immersion in the UK [cf. Athanasopoulos, 2009, Bilingualism: Language and Cognition, 12, 83-95].
A method to study colour category
If there are perceptual colour categories which are not reduced to the verbal categories, then a problem is how to look into these perceptual categories not resorting to verbal names, labels and the like. I suggest to use a method which is based on the same idea as the partial hue-matching technique developed recently. The results of some preliminary experiments will be reported.
How invariant is unique white?
Sophie Wuerger, Kaida Xiao, Emily Hird, Tushar Chauhan, Dimos Karatzas and Esther Perales
Despite the theoretical importance of unique white, there is little agreement on its precise chromaticity. Often an equal-energy white (CIE x=0.33; y=0.33) is assumed [Werner & Shiffrin, 1993, JOSA, 10(7), p.1509-1516.] which is close to ecologically relevant illuminations, such as the sun’s disk (x= 0.331; y = 0.344) and daylight (D65: x = 0.313; y = 0.329). Here we test the invariance of these unique white settings under changes in illumination, task and luminance. Stimuli were displayed on a CRT on a black background and ambient illumination was controlled by a Verivide luminaire. White settings were obtained (n=30) under dark viewing conditions, under D65 (x= 0.312 y=0.334), and under CWF (x=0.394 y=0.387), using three different tasks: adjustment along the daylight locus, along the axes in LUV space, or along the unique hue lines. We find that the average unique white point (under dark viewing conditions) is located at CIE x=0.292, y=0.303, which is at a significantly higher colour temperature than daylight. Changing the illumination from dark to D65 (CWF) shifted the white point towards D65 (CWF). We conclude that observers are able to provide accurate but illumination-dependent unique white settings. Implications for different adaptation models will be discussed.
Category effects for red and brown
Christoph Witzel and Karl Reiner Gegenfurtner
Red and brown are particular colour categories: Their member colours are comparatively dark and change category membership with increasing lightness to orange and pink, respectively. Moreover, brown is neither a unique nor a binary hue, and seems to be only defined through language. Brown also appears much later during colour term acquisition. We investigated category effects for the red-brown category boundary. We established the red-brown boundary through a naming task, measured discrimination thresholds for colours across the boundary, and performance in a visual search task with colour pairs that were equalised in discriminability based on the empirical discrimination thresholds. We found that there is no change of discrimination thresholds at the boundary. In contrast, there was a boost of performance (lower reaction times, accuracy twice as high) for identifying colour differences in equally discriminable colour pairs, when the colours cross the boundary. These category effects were not lateralised at all. These results are completely in line with those shown for colours at moderate lightness levels. Given the particularity of brown, these results further underpin the idea that category effects are due to a shift of attention to the linguistic distinction between categories rather than being a pure product of perception.
Locating colors in the Munsell space: an unconstrained color naming experiment
Giulia Paggetti and Gloria Menegaz
A previous study [Paggetti et al., 2011, Attention, Perception and Psychophysics, 73(2), 491-503] based on a constrained color naming experiment on Italian subjects suggested the need of a twelfth basic color term (BCT) within the blue category. Though, it is still controversial whether constraining the subject’s answers would introduce a bias on the subject's performance and thus lead to erroneous conclusions. For this reason, a second color naming experiment was performed following the unconstrained method. In order to overcome some limitations of the OSA-UCS system used previously, the Munsell system was adopted. The two main objectives of this work were to identify color classes and color names during an unconstrained color naming task and to compare the outcomes with those obtained following the constrained method. Two sets of measures were extracted for characterizing each color term (consistency and consensus) and color category (centroid and focal colors). Results support the conclusions driven from the previous study suggesting that the Italian language features twelve BCTs. This study contributed to identify color classes as defined by Italian speakers during unconstrained color naming, as well as to the definition of the positions of focals, centroids, consistency and consensus colors in the Munsell system.