[ARTICLE] The psychology of "cuts"

[purplepolecat]

An interesting article about how our brain processes movie cuts. It seems that our visual perception is already slicing things up into discrete scenes, so watching edited cuts on the screen is not significantly different from what we experience in real life.
http://aeon.co/magazine/psychology/why- ... ovie-cuts/

"These examples reinforce the story about why most film editing feels natural and unobtrusive. When the editing supports the way our visual systems evolved to function, it tends to recede into the background of our ongoing visual functioning. This suggests that filmmakers must be pretty good intuitive perceptual psychologists."

I would imagine that this is also true for AMVs. We instinctively know (or think we know) when a cut feels right, and when it feels jarring. Syncing with music adds another layer of complexity at the editing level, but can also help the brain process the images smoothly by providing cues.

[snapxynith]

I saw three important elements that were touched in that article. Change blindness, scene change detection, scene continuity detection between cuts were the noted mental structures.

Feeling out a cut using those systems let's us do two things. Making scenes play out as if they were actually in the same location telling a slice of a story and breaking scene connectivity to start a new one by adding enough unrelated data.

This innate skill that our minds have is very apparent especially in young AMV editors and was only made more apparent by comparison by random clip constructions made by our robot editor Toby (https://www.youtube.com/user/tob39amv). Toby does random clip selection over any requested source material, with a cut length that is determined before the render. What happens in many of his videos seems relatively consistent if not a little boring at times. What the videos he makes do not feel is fully random. Many of his videos have enough emergent patterns without context that the minds of most polled audiences are willing enough to add their own context. Leaving his videos to have an unexpected narrative from the source material. There is another oddity with Toby. His videos have a property that is very different from young editors. None of them ever seem to touch the realm of "cringe worthy" editing.

You know the feeling, that irksome dissonance, the clashing of emotional content. Young editors show how good they are at picking context and narrative that are important. Yet, how badly they consciously understand what context they are putting up on screen. They tend to pick scenes that are emotionally rich, context heavy, and slam them right up against another scene that doesn't enhance the other but instead fights against it. Similarly, we see the scenes telling a radically different message than the underlying music causing a rift between them. An unsettling type of pattern that roughs us up mentally. Making each second a conflicting bucket of emotions fighting for dominance. A communication breakdown.

So as editors grow and consciously gain control of that power to select scenes with and without continuity, we see more consistent storytelling, communication, and emotional matching to their full message. Whereas Toby usually caps out at around 2-4 accidental correlations per video.

This article manhandled a paper I have reflected on for a long time after its publication. http://people.psych.cornell.edu/~jec7/p ... hsci10.pdf
If there's one thing a lot of people get out of it, it's "cuts are getting shorter." Which as I understand it is not quite the case. Cuts are getting more in line with a specific pattern. The 1/f pattern. So if a cut was long, there could only be so many of them, so you see less of that long style cut but also see more medium length cuts in proportion. With short cuts being more numerous but in a relative proportion to the content filled long cut.

In this paper's scenario the measurement is of two things: cut length (power), number of cuts with a specific length (frequency).
It shows there's a distinct pattern that has been selected over time by audiences. A pink noise pattern. There are cuts that are long and cuts that are short but the frequency that each relatively shows up in a movie is inversely related to their length.

So the longer the cut, the less you'll see cuts of that length added but only because a cut of that length already existed. The smaller the cut the higher the chance you'll see a lot more of them up to a specific limit. Popular movies tended to have a distribution of those lengths balanced around the 1/f curve. Meaning that movies that relentlessly crushed you with cut after cut in short order, were rough on us, taxing our attention and our ability to keep up. Movies with looooong shots that never ended would lose our attention more often. The best were the movies that balanced their pacing in this specific way. So that information rich shots stayed longer but were rare in comparison and small cuts conveyed little information for a specific time vs the total length of the work.

Showing us that as people, as an audience, we look for that balance. Not to rush us, not to linger. Tell us a variable paced story, it keeps our attention the best.

[dokidoki]

Here's a graph Wired made of cuts/minute in movie trailers over the years. (Dr. Strangelove is an outlier at 136 )
http://www.wired.com/2013/06/online-trailers-cuts/

snapxynith: That's interesting about Toby, though I don't have much time to check out the videos at the moment.