Evidence that “disfluencies” in speech enhance comprehension
nautil.us
nautil.us
I recall showing my mom Paul Buchheit's Startup School 08 talk:
https://www.youtube.com/watch?v=EZxP0i9ah8E
I thought it was great, and I remember listening very intently to parse out exactly what he was saying and figure out what he was saying. I showed it to my mom, and her first reaction was "Why are you listening to him? He seems kinda nebbishy."
Notice how many pauses, umms, y'knows, and other filler words are in that speech. I ignored them because I already knew what he had accomplished; my assumption was that he knew what he was talking about, and so I should pay close attention. My mom, who knows nothing about tech and didn't start using GMail until 2012, came away with the impression that he was unconfident, rambling, and didn't really know what he was talking about.
The advice to eliminate filler words may be a consequence of a world where people make snap judgments about the credibility of a person in seconds, and usually are not qualified to judge the content of the speech at all. Notice how the article says that this advice first caught on in the 20th century, right as mass communication and soundbite messages caught on. The studies they cite, however, all featured listeners from the same field as the speaker who already cared about the content.
Obviously there are variations on use; where cases like overuse from being unprepared can be painful to listen to... But when obviously pausing to think of the best way to describe something to a particular/unfamiliar audience, or summarize a broader idea down to a specific area being discussed, I've noticed how helpful it can be in understanding, at least for the way that /I/ listen to people.
When I hear somebody "um", to think of the best intersection between what they know, context, and what I know, my mind is trying to do the exact same thing, or even beat them to it. For me, this definitely engages me more; even if I come to the wrong conclusion, my conscious thought is saturated with putting these pieces together. Any related information is already active, and easier to link together. -- Compared to someone speed-reading a prepared statement, where I may come away with some high level knowledge, but be too busy parsing the flow of speech to contemplate many of the finer points as they come.
I'm sure most of us could come up with some examples in short order though. Here's something close: saying "dead XXX" like "dead simple". We know that when we hear that it means something like "really simple" by why would "dead" work there? Of course English is famous for overloading words with many meanings, and this stuff is dead common. But I still find it interesting to ponder the various phrases and what-not that we use.
None of which, of course, has a ton to do with the material in this article. It was really just the title that got me thinking down this path...
I was not able to find the right term for "orphaned" words like dander. Does anyone know what the term is?
Indeed; anyone with an allergy is well familiar with pet dander; both the word and the "stuff".
"Raining cats and dogs", "by the skin of one's teeth", "steal someone's thunder".
Regarding the article, I thought it was going to allude to things like "implicit microaggressions". Was glad to see it was focused more on credibility and use of "uhm", "ah" and other natural stumbling blocks of speech.
The more someone pauses, to me, the more sincere they appear to me. As if they are trying to think of the best way to phrase something for their audience. Even if in reality they're frantically trying to remember what they were about to say...
See? Can similes be colourful? Can speech be spicy?
To me this chimes with human brains being pattern seeking/matching machines. Enough meaning leaks across discrete word definitions that we can freely use phrases such as "an icy stare", "a prickly relationship" to appeal to senses that are not directly involved in stares or relationships.
I have heard language theorists suggest that this appeal to multi-sensory communication may have influenced the evolution of language among early humans. Unfortunately I have no memory of who said this.
Still. I enjoy the theory.
Corollary: removing disfluencies from online lectures diminishes comprehension.
I took Ullman's "Automata" course, and they'd edited out all the um's and pauses etc in the vidoes, making them half as long. It reduces download size, but made them really difficult to understand. You don't have time to process one interesting thought before the next is upon you. Very different from live lectures. I often had to rewind 5 to 10 times.
Yes, the material isn't easy and Ullman is full of insights, but this happened even with the beginning of the introductory lecture.
From this article, it seems that the um's and pauses would also have helped signal difficult vs easy ideas. (Sometimes I worked hard to grasp his precise meaning, only to realize he was speaking casually).