Abstract
Longer processing times are often associated with a larger attentional demand. Using the inattentional blindness paradigm (IB) by Mack and Rock (1998), Moore, Grosjean, and Lleras (2003) provide an operational definition of attention that may not be consistent with this idea. In their studies, a main task can be affected by a background pattern, and in a critical trial it is examined whether the participant is inattentionally blind to the background. If the participant cannot report the background but the main task is still affected by it, then it is said to be processed inattentively. Although the maximum presentation time is not fixed by this definition, Moore et al. used short ones. In this study, however, we show that some processes are inattentive and nevertheless require a long presentation time. In Experiments 1 and 2, we examined whether line length judgments were affected by a background of differently oriented Gabor patches. These patches could sometimes be oriented in such a way that they formed an upright or an inverted V (i.e., the two tracks in the Ponzo illusion) based on texture segregation and were otherwise randomly oriented. In a critical trial the participant was asked whether he or she had seen the background pattern, and to guess whether it showed an upright or an inverted V, along with a confidence rating of it. Results showed that the illusion could be observed inattentively with a 500 ms but not with a 200 ms presentation time. In Experiment 3, we used a letter discrimination task and a background of horizontal Gabor patches. One patch, however, was vertical and appeared either to the left or the right of the letter, either compatible or incompatible with the required response to the letter. Even with a 500 ms presentation time, this task could not be performed inattentively, demonstrating the validity of our procedure in Experiments 1 and 2. We conclude that, using Moore et al.'s definition, attention is independent of processing time.