Independent effects of colour on object identification and memory

Abstract

We examined the effects of colour on object identification and memory using a study–test priming procedure with a coloured-object decision task at test (i.e., deciding whether an object is correctly coloured). Objects were selected to have a single associated colour and were either correctly or incorrectly coloured. In addition, object shape and colour were either spatially integrated (i.e., colour fell on the object surface) or spatially separated (i.e., colour formed the background to the object). Transforming the colour of an object from study to test (e.g., from a yellow banana to a purple banana) reduced priming of response times, as compared to when the object was untransformed. This utilization of colour information in object memory was not contingent upon colour falling on the object surface or whether the resulting configuration was of a correctly or incorrectly coloured object. In addition, we observed independent effects of colour on response times, whereby coloured-object decisions were more efficient for correctly than for incorrectly coloured objects but only when colour fell on the object surface. These findings provide evidence for two distinct mechanisms of shape–colour binding in object processing.

Keywords:

• Object
• Colour
• Identification
• Priming
• New-association priming
• Memory

This work was supported by funding from the Wales Institute of Cognitive Neuroscience and the Leverhulme Trust to Toby Lloyd-Jones.

Notes

¹ Luminance values in Photoshop (L*) correspond to the “luma” component of the image—that is, the weighted sum of gamma corrected red, green, and blue components of the colour signal (Poynton, 2003; Stone, 2004). Luma is calculated using Rec. 601 coefficients: Rec. 601: Luma (Y′) = 0.299 R′ + 0.587 G′ + 0.114 B′. The term luma was intended to prevent confusion between luma as implemented in video engineering and luminance as used in colour science. There is no associated scientific unit of measurement of luma as currently there is no conventional numerical description for this stimulus characteristic.