We develop an image-level analysis of errors, which shows coarse location information exists in many apparently ‘incorrect’ location responses. Here, we explore what type of information is present in this timeframe, particularly focusing on the presence (or lack thereof) of information about the location of potential abnormalities. This implies that the information extracted from the image in the first second of processing contains critical information that informs diagnosis. With time and practice, experienced radiologists are thought to develop skills that allow them to form the basis of a diagnosis (normal or abnormal) during an initial glance at an image. is that a cancer or not?), often under time pressure. In medical imaging, a radiologist searches and interprets a medical image to make critical diagnostic decisions (e.g. The effect of density on detection and localisation highlights the importance of considering breast density in medical screening. We emphasise the importance of taking these factors into account when interpreting results. Although there may be a dissociation between detection and localisation on a small proportion of trials, we find a number of factors that lead to the underestimation of localisation including stimulus variability, response imprecision and participant guesses. We found both detection and localisation information for all conditions. Participants were asked to decide whether there was an abnormality (detection) and then to locate the mass on a blank outline of the mammogram (localisation). As the female breast varies with respect to the level of normal fibroglandular tissue, the images were categorised into high and low density (50% of each), resulting in difficult and easy searches, respectively.

Extending previous work, we presented radiologists with unilateral mammograms, 50% containing a mass, for 250 or 1000 ms. Although radiologists have been shown to be able to detect an abnormality ‘above chance’ at short durations, the extent to which abnormalities can be localised at brief presentations is less clear. The ability of an experienced radiologist to rapidly detect an abnormality on a mammogram may build upon this general capacity. Humans can extract considerable information from scenes, even when these are presented extremely quickly. Chance is 4.4% and adjusted to 9.1% when including the ROA (dotted line) with 95% confidence intervals. The three radiologists that had piloted the experiment previously are illustrated in red. Detection and localisation results: percentage correct on the localisation task when a region of acceptance (ROA) around the lesion is included for individual radiologists for (a) low-density and (b) high-density mammograms.

Detection and localisation results: percentage correct on the localisation task for individual radiologists on trials when detection was correct for (a) low-density and (b) high-density mammograms. Detection accuracy: sensitivity (d′) for individual radiologists for (a) low-density and (b) high-density mammograms. Detection accuracy: percentage correct for individual radiologists on target absent trials for (a) low-density and (b) high-density mammograms on the detection task. Detection accuracy: percentage correct for individual radiologists on target present trials for (a) low-density and (b) high-density mammograms on the detection task.

Graphicconverter 9 mouse clicks