WEBVTT 00:00:15.040 --> 00:00:17.640 Take a series of still, sequential images. 00:00:18.360 --> 00:00:19.800 Let's look at them one by one. 00:00:23.760 --> 00:00:24.960 Faster. 00:00:28.440 --> 00:00:30.256 Now, let's remove the gaps, 00:00:30.280 --> 00:00:31.480 go faster still. 00:00:32.920 --> 00:00:34.440 Wait for it ... 00:00:36.160 --> 00:00:37.576 Bam! 00:00:37.600 --> 00:00:38.800 Motion! 00:00:39.520 --> 00:00:40.976 Why is that? 00:00:41.000 --> 00:00:44.536 Intellectually, we know we're just looking at a series of still images, 00:00:44.560 --> 00:00:46.496 but when we see them change fast enough, 00:00:46.520 --> 00:00:50.096 they produce the optical illusion of appearing as a single, persistent image 00:00:50.120 --> 00:00:52.360 that's gradually changing form and position. 00:00:53.000 --> 00:00:56.456 This effect is the basis for all motion picture technology, 00:00:56.480 --> 00:00:58.176 from our LED screens of today 00:00:58.200 --> 00:01:01.056 to their 20th-century cathode ray forebearers, 00:01:01.080 --> 00:01:04.215 from cinematic film projection to the novelty toy, 00:01:04.239 --> 00:01:07.096 even, it's been suggested, all the way back to the Stone Age 00:01:07.120 --> 00:01:09.120 when humans began painting on cave walls. 00:01:09.760 --> 00:01:13.016 This phenomenon of perceiving apparent motion in successive images 00:01:13.040 --> 00:01:15.416 is due to a characteristic of human perception 00:01:15.440 --> 00:01:18.576 historically referred to as "persistence of vision." 00:01:18.600 --> 00:01:19.816 The term is attributed 00:01:19.840 --> 00:01:22.816 to the English-Swiss physicist Peter Mark Roget, 00:01:22.840 --> 00:01:24.456 who, in the early 19th century, 00:01:24.480 --> 00:01:27.376 used it to describe a particular defect of the eye 00:01:27.400 --> 00:01:28.976 that resulted in a moving object 00:01:29.000 --> 00:01:31.680 appearing to be still when it reached a certain speed. 00:01:32.400 --> 00:01:35.696 Not long after, the term was applied to describe the opposite, 00:01:35.720 --> 00:01:37.736 the apparent motion of still images, 00:01:37.760 --> 00:01:41.520 by Belgian physicist Joseph Plateau, inventor of the phenakistoscope. 00:01:42.280 --> 00:01:46.456 He defined persistence of vision as the result of successive afterimages, 00:01:46.480 --> 00:01:49.216 which were retained and then combined in the retina, 00:01:49.240 --> 00:01:52.936 making us believe that what we were seeing is a single object in motion. 00:01:52.960 --> 00:01:55.856 This explanation was widely accepted in the decades to follow 00:01:55.880 --> 00:01:57.976 and up through the turn of the 20th century, 00:01:58.000 --> 00:02:01.296 when some began to question what was physiologically going on. 00:02:01.320 --> 00:02:04.336 In 1912, German psychologist Max Wertheimer 00:02:04.360 --> 00:02:06.976 outlined the basic primary stages of apparent motion 00:02:07.000 --> 00:02:09.256 using simple optical illusions. 00:02:09.280 --> 00:02:11.096 These experiments led him to conclude 00:02:11.120 --> 00:02:14.920 the phenomenon was due to processes which lie behind the retina. 00:02:15.640 --> 00:02:17.616 In 1915, Hugo Münsterberg, 00:02:17.640 --> 00:02:20.056 a German-American pioneer in applied psychology, 00:02:20.080 --> 00:02:22.936 also suggested that the apparent motion of successive images 00:02:22.960 --> 00:02:25.496 is not due to their being retained in the eye, 00:02:25.520 --> 00:02:28.440 but is superadded by the action of the mind. 00:02:29.520 --> 00:02:32.096 In the century to follow, experiments by physiologists 00:02:32.120 --> 00:02:34.496 have pretty much confirmed their conclusions. 00:02:34.520 --> 00:02:36.856 As it relates to the illusion of motion pictures, 00:02:36.880 --> 00:02:39.576 persistence of vision has less to do with vision itself 00:02:39.600 --> 00:02:41.856 than how it's interpreted in the brain. 00:02:41.880 --> 00:02:45.416 Research has shown that different aspects of what the eye sees, 00:02:45.440 --> 00:02:48.696 like form, color, depth, and motion, 00:02:48.720 --> 00:02:51.296 are transmitted to different areas of the visual cortex 00:02:51.320 --> 00:02:53.216 via different pathways from the retina. 00:02:53.240 --> 00:02:54.736 It's the continuous interaction 00:02:54.760 --> 00:02:56.856 of various computations in the visual cortex 00:02:56.880 --> 00:02:59.136 that stitch those different aspects together 00:02:59.160 --> 00:03:01.056 and culminate in the perception. 00:03:01.080 --> 00:03:02.936 Our brains are constantly working, 00:03:02.960 --> 00:03:05.696 synchronizing what we see, hear, smell, and touch 00:03:05.720 --> 00:03:07.016 into meaningful experience 00:03:07.040 --> 00:03:09.616 in the moment-to-moment flow of the present. 00:03:09.640 --> 00:03:12.816 So, in order to create the illusion of motion in successive images, 00:03:12.840 --> 00:03:14.856 we need to get the timing of our intervals 00:03:14.880 --> 00:03:17.720 close to the speed at which our brains process the present. 00:03:18.800 --> 00:03:21.936 So, how fast is the present happening according to our brains? 00:03:21.960 --> 00:03:23.176 Well, we can get an idea 00:03:23.200 --> 00:03:25.656 by measuring how fast the images need to be changing 00:03:25.680 --> 00:03:26.896 for the illusion to work. 00:03:26.920 --> 00:03:29.896 Let's see if we can figure it out by repeating our experiment. 00:03:29.920 --> 00:03:33.976 Here's the sequence presented at a rate of one frame per two seconds 00:03:34.000 --> 00:03:36.416 with one second of black in between. 00:03:36.440 --> 00:03:37.656 At this rate of change, 00:03:37.680 --> 00:03:39.776 with the blank space separating the images, 00:03:39.800 --> 00:03:41.936 there's no real motion perceptible. 00:03:41.960 --> 00:03:44.256 As we lessen the duration of blank space, 00:03:44.280 --> 00:03:47.016 a slight change in position becomes more apparent, 00:03:47.040 --> 00:03:49.496 and you start to get an inkling of a sense of motion 00:03:49.520 --> 00:03:50.920 between the disparate frames. 00:03:51.480 --> 00:03:52.720 One frame per second. 00:03:55.520 --> 00:03:56.800 Two frames per second. 00:03:59.480 --> 00:04:00.920 Four frames per second. 00:04:02.440 --> 00:04:04.616 Now we're starting to get a feeling of motion, 00:04:04.640 --> 00:04:06.576 but it's really not very smooth. 00:04:06.600 --> 00:04:09.816 We're still aware of the fact that we're looking at separate images. 00:04:09.840 --> 00:04:11.760 Let's speed up. Eight frames per second. 00:04:14.200 --> 00:04:15.680 12 frames per second. 00:04:16.560 --> 00:04:18.240 It looks like we're about there. 00:04:21.440 --> 00:04:24.936 At 24 frames per second, the motion looks even smoother. 00:04:24.960 --> 00:04:26.760 This is standard full speed. 00:04:28.120 --> 00:04:30.816 So, the point at which we lose awareness of the intervals 00:04:30.840 --> 00:04:32.416 and begin to see apparent motion 00:04:32.440 --> 00:04:35.120 seems to kick in at around eight to 12 frames per second. 00:04:36.040 --> 00:04:38.776 This is in the neighborhood of what science has determined 00:04:38.800 --> 00:04:40.896 to be the general threshold of our awareness 00:04:40.920 --> 00:04:42.216 of seeing separate images. 00:04:42.240 --> 00:04:44.656 Generally speaking, we being to lose that awareness 00:04:44.680 --> 00:04:47.056 at intervals of around 100 milliseconds per image, 00:04:47.080 --> 00:04:50.216 which is equal to a frame rate of around ten frames per second. 00:04:50.240 --> 00:04:51.616 As the frame rate increases, 00:04:51.640 --> 00:04:53.776 we lose awareness of the intervals completely 00:04:53.800 --> 00:04:56.720 and are all the more convinced of the reality of the illusion.