Visual distraction in cytopathology: should we be concerned?

Abstract

Visual distraction in cytopathology has not been investigated previously as a source of diagnostic error, presumably because the viewing field of a conventional light microscope is considered to be large enough to minimise interference from peripheral visual stimuli. Virtual microscopy, which involves the examination of digitised images of pathology specimens on computer screens, is beginning to challenge the central role of light microscopy as a diagnostic tool in cytopathology. The relatively narrow visual angle offered by virtual microscopy makes it conceivable that users of these systems will be more vulnerable to visual interference. Using a variant of a visual distraction paradigm (the Eriksen flanker task), the aim of this study was to determine whether the accuracy and speed of the interpretation of cells on a central target screen are affected by images of cells and text displayed on neighbouring monitors under realistic reading room conditions. Following a brief period of training, 31 cytology novices undertook four cell interpretation tests under different conditions of visual distraction. Error rates were measured under each condition. There was no effect of visual distraction on diagnostic accuracy. To the extent that the results from cytology novices extend to experienced practitioners, visual distraction is an unlikely source of error in virtual microscopy. Efficient visual selection and spatial attention, coupled with the high perceptual load of target images and the peripheral location of distractors, provide plausible explanations for the observed results.