Cortical responses to invisible objects in the human dorsal and ventral pathways

Design: fMRI requires us to record data for a base case and then record again when the concerned task is being performed. The activation pattern obtained in both cases is subtracted to detect the brain regions critical to that task. The present study was interested in knowing the activation pattern associated with objects. Hence, they chose a base case of the same object being scrambled and non recognizable. The study was also interested in the difference in activation when the object was perceptually visible vs perceptually invisible (suppression in binocular rivalry). They ran separate scans for the visible and the invisible conditions. In each scan they alternated between 20s durations of scrambled input and 20s durations of unscrambled input. In each 20s run, the subject was shown 40 objects. Each object was shown for 200ms, followed by a fixation task for 300 milliseconds. To make the objects visible in one case, the rivalling stimuli was taken as a blank screen. To make the object invisible in the other case, the rivalling stimuli was taken as a bright red noise. Higher contrast and brightness makes the stimuli stronger against the rivalling stimulus and can dominate perception for minutes at a time. However, this does not negate the possibility of the rival overcoming suppression during the experiment. To identify these cases two methods where used. One was to simply ask the subjects. The second was to conduct an 2AFC (two alternative forced choice) and note whether the subjects were performing above chance level in detecting whether the first 20s was scrambled or the next. Note: Since, the random noise in the invisible case is not an object, it does not significantly influence the activation in the region of interest. If this were not the case, one could assert that the observation made during the invisible state were relating to the random noise than the invisible stimuli.