parallel processing

Meaning: Parallel processing is the simultaneous handling of several kinds of information by separate neural pathways, in contrast to serial processing, which works one step at a time. In vision, color, form, motion, and depth are analyzed at the same time and then combined into one percept. Automatic tasks, such as recognizing a familiar face, rely heavily on it, while controlled, effortful tasks tend to be more serial. In connectionist models, parallel distributed processing means knowledge is spread across networks of linked units that are active at once.

On the MCAT: A visual search passage compares response times. In a feature search (one red item among green ones), time changes little as distractors are added, suggesting parallel processing. In a conjunction search (a red X among red O's and green X's), time rises with the number of distractors, suggesting serial processing. This is the logic of feature integration theory.

Common confusion: Parallel versus serial describes timing, not direction. Bottom-up versus top-down describes whether processing is driven by the stimulus or by knowledge and expectations. Parallel processing is also not the same as multitasking. The brain processes many inputs at once, yet dividing attention between two demanding tasks usually lowers performance.