ABSTRACT
In normal circumstances, our ability to put names to objects is strikingly efficient. We can find the name for perhaps some thousands of objects within only a few seconds, and name retrieval for familiar objects can take place well within that time. Yet, in naming an object, the brain needs to pass through a number of processing stages. To begin with, early visual processes must encode the shape and possibly also the surface details of the object (its colour, texture, and so forth). After this, the encoded perceptual information must be matched to memory. Different forms of stored memory may also need to be accessed: knowledge about the form of the object (its stored structural description), about its functional and associative properties (its semantic description), and its name (perhaps in an abstract form, perhaps even its phonological description). Access to the various knowledge forms can be thought of as constituting different stages in the naming process. Here we ask how the “object” part of object naming-including the processes involved in accessing structural and semantic memories-constrains the “naming” part-the processes involved in finding a unique name and eventually a unique phonological description for a stimulus. Is the stage of name
retrieval in object naming the same as that involved in naming other stimuli (printed words, descriptions, sounds), or are name retrieval processes fashioned by the nature of preceding visual recognition processes required for objects? To answer such questions, we will first consider how the processing stages in object naming might operate, before going on to evaluate how visual recognition processes impact on name retrieval. Our answers are based on converging evidence drawn from experimental and neuropsychology, functional imaging, and computational modelling.
