Engaging in Navon processing prior to face recognition influences recognition performance (Macrae & Lewis, 2002; Perfect, 2003; Weston & Perfect, 2005). Past research has shown that global Navon processing can improve face recognition accuracy whereas local Navon processing impairs it. One explanation, and the dominant one in the literature, is that the Navon letter task elicits holistic and featural processing styles which map onto the holistic and featural processing used in face recognition. However, this theory has not been extensively tested in the literature. This thesis investigates whether processing style can explain the effects caused by the Navon letter task, examining the effects across a variety of recognition tasks such as the composite face task (Young, He'llawell & Hay, 1987) and the change detection task (Leder & Bruce, 2000). In addition, the effects of Navon processing were examined across different recognition tests such as forced choice, old / new and matching presentation formats. Furthermore, the processing style account of Navon processing is based on the Transfer Appropriate Processing (TAP) account of memory (Roediger, 1990). With regards to face recognition this theory suggests that performance depends upon the similarity between encoding and retrieval processes. Therefore, the experiments presented in this thesis also examined the influence of Navon processing on retrieval from memory following manipulations at encoding. The results of nine experiments showed that the effects of Navon processing were not consistent across all experimental situations and that the magnitude of the effect depends upon a number of factors such as the encoding task used, the task used at retrieval and the style of the Navon letters. Whilst some results were consistent with the processing style theory, other findings were not. Therefore, alternative explanations based on spatial frequency and hemispheric asymmetries are provided for the Navon effect.

Document Type


Publication Date