Abstract
Blur has previously been suggested to be processed preattentively (Loschky et al., 2014). Blur may be responded to in a similar fashion to other preattentive features such as color, orientation, size, etc., which can capture attention leading to 'pop out' (Treisman & Gelade, 1980). However, previous eye-tracking studies have found that when blur/clarity contrast exists in an image, eye-movements are typically directed towards regions of clarity and ignore blurred regions (Enns & MacDonald, 2012; Kahn, Dinet, & Konik, 2011; Loschky & McConkie, 2002; Smith & Tadmor, 2012). Peterson (2016) showed in a rotated L and T visual search task that uniquely blurred letters were ignored by selective attention, but uniquely clear items captured attention, as shown by reaction time results. The current study added eye-tracking to the methods used in Peterson (2016) to investigate if the previous reaction time results were an effect of unique blur being either ignored or repelling attention at the start of a rotated L versus T visual search. Set sizes of four and eight letters were used. Letters were presented on an imaginary circle of eight degrees radius. Uniquely blurred and clear singleton letters were manipulated to be nonpredictived of target location. Similarly to the results of Peterson (2016), reaction times at both set sizes (4 & 8) suggested that blur is ignored by selective attention. However, investigation of the first letter fixated in the visual search showed that uniquely clear letters strongly captured attention, while uniquely blurred letters weakly repelled attention at each set size. These findings suggest that overall reaction times are not sensitive enough to detect the weakly repulsive nature of blur to selective attention, but with eyetracking we have found evidence to suggest that unique blur repels, though not nearly as strongly as attention is captured to uniquely clear letters.
Meeting abstract presented at VSS 2017