Abstract
Researchers may seek to manipulate visual recognition while maintaining the the underlying low-level visual properties of an image. To this end, researchers have used a class of stimuli known as “texforms” that are unrecognisable yet closely match the low-level properties of the original, recognizable stimuli from which they are synthesized. Texforms, however, can require a significant amount of time to generate using parameters specified by previous researchers, thereby limiting one’s ability to conduct studies that require many exemplars. Furthermore, if a researcher seeks to manipulate the recognisability of texforms to varying degrees, the computational processing time can increase even further. By using different parameters, we show that one can synthesize texforms much more quickly, which permits the generation of many exemplars at different degrees of recognisability. We also compare the low-level visual properties of the texforms generated with different parameters.