RESEARCH NOTES #2: All filler? No thriller...

Our new research focuses on the effects of filler words - on us, and our audiences - and what we can do to eradicate them

RECENTLY I was walking through Manhattan on the way to see a client and three students from New York University passed by me, with the person nearest talking to the other two.

In the time he was audible - amid traffic noise and general hubbub, and he wasn't speaking that loudly at all - I noticed him use the word 'like' on five separate occasions. In the space of maybe four seconds.

If you've ever eavesdropped on a Gen Z conversation, you'll know this can be par for the course. I'm not criticising - it’s just the lingua franca of an in-group, no matter how infuriating it may be once you've noticed that one word permeating the discourse, and can't stop noticing.

Because we all have them. The words we stick in, for no real reason. Maybe even just the 'um' and 'err' of an unprepared verbal narrative.

I was dictating some work into Word recently when a client called with something they wanted a steer on. The dictate function stayed on. Twenty minutes later I ended the call and saw the flow-of-consciousness copy added to my document in error. One phrase popped out: 'y'know'. Twelve times.

Why do we do it? When do audiences get tired of it, and start making judgements, such as 'he's lost his train of thought' or 'she hasn't prepared', or 'they're not sure'? Or, is it actually just one way that we sound more conversational and human, as opposed to sounding like that other lovely Word function, 'Read Aloud'?

Spontaneous speech is rarely fluent. It’s often filled with pauses in the form of filler words that are usually associated with the speech-planning process. Most filler words occur due to issues with lexical searching - trying to find the right words - which is influenced by several factors and causes issues in comprehension and clarity (Oomen and Postma, 2001).

Filler words can be defined as “any word that interpolates the main message of the speaker”. Which, as we all know from experience, includes words and utterances like um, like, uh or ya know (Duvall et al., 2014)

The problem with interpolation is it has an impact on the credibility and clarity of the speaker, and on the comprehension of the listener. If we want to eradicate or minimise the fillers, it’s important to understand their causes, and their negative (and even sometimes positive) impacts on both the listener and the speaker, so that we can communicate more effectively (Duvall et al., 2014).

What are the causes and why do speakers ‘fill the void' with fillers? If you know the conditions under which they most often occur, you can start to deal with them

If these factors are in the mix, then the speaker is more likely to use filler words:

1. Infrequent words. When using infrequent words which we don’t deploy on a daily basis, it can be difficult to process in our mental dictionaries. When unable to locate a word, a filler word may be thrown in, in place of the word. (Duval et al., 2014)

   
   

2. Nervousness and anxiety. Glossophobia is fear and anxiety associated with public speaking. This also links to nervousness and can result in speakers wanting to talk quickly, and finish their speech quickly. This urge to get it over with creates an inability to properly process speech, and then finding the next word can be a challenge. As a result, we use a filler word, which is an easier route than identifying the next word or phrase (Seals and Coppock, 2022).

3. Preparation time. Without proper prep time, statement-to-statement transitions are not effectively established which creates uncertainty in speech which again leaves the speaker seeking fillers when trying to remember their next point (Seals and Coppock, 2022).

4. Divided attention. If the speaker is trying to focus on too many things at once, such as a distracting audience member or a flickering light, then distraction from the speech can see us drafting in filler words (Duval et al., 2014)

Fillers diminish our credibility as speakers. Conrad et al., (2012) conducted a study that recorded the successful conversion rates of telemarketers. The study explored a group of telemarketers and gauged those who used more, or fewer, fillers. The success rates dropped for those who used more, perhaps not surprisingly. But the hurdle wasn’t high; the success level was especially impacted if the number of fillers in the interaction rose above 1.28 words per hundred.

Furthermore, when Caroline Kennedy ran for New York Senate, in an interview she used an average of 12.66 filler words per minute. She was slated, receiving extensive criticism in newspapers with one comment, probably burying her senate ambitions, citing, ‘cringing verbal tics... showed her inexperience’ (Duval et al., 2014).

When it really matters, as opposed to casual conversation, fillers can cause impairment to how we actually comprehend language.

A study on Japanese students learning English found that filler words caused distinct comprehension issues when learning the language. Misinterpretation of the filler word led to processing errors, and depending on where the filler word was placed, changed comprehension among the students. (Carney, 2022)

The way in which we communicate so effortlessly as human beings, in spontaneous dialogues like conversations, is partly determined by how we can predict what the order flow of words will be. We’ve all seen predictive text on our phones; sometimes it even works. Imagine how good our brains are at expecting certain things to come next.

But another study found that when subjects were given basic tasks, and where some subjects had received instructions that contained more filler words, it led to less or incorrect understanding of the task. In a similar study, participants were asked to predict what would follow filler words after they had occurred in spontaneous language. Participants who were in the filler word group more often predicted the incorrect following word, compared to those who had no filler word, and they had better understanding than the ‘filler group’. (Duval et al., 2014).

So, we know that behaviours like going too fast; not having prepared links between our ideas; reaching for unfamiliar words; and being distracted, all increase the use of filler words, with the impact it has on others’ perception of us, and their comprehension of what we are trying to communicate.

How do we stop these fillers having a negative impact on our ability to engage with others?

The first is awareness - making a speaker aware of their use of filler words. Well-meaning colleagues might mention it to you after a speech, but quite frankly they may be reluctant to bring up the problem.

I have no such qualms. My problem is, people don’t believe you. So you have to show them the playback and it’s often quite a revelation, like a snorer being presented with the evidence of their misdemeanour. “But I didn’t even hear myself doing it!” is the usual response.

Seals and Coppock (2022), aimed to improve their mentees’ speech by getting them to calculate the percentage of filler words to total words that contain information and ask mentees to identify non-filler related content in speech. Results were ‘remarkable’ with uniform improvements in oral presentation which clear improvement in presentation skills.

Our research dug up a piece in Harvard Business Review where the authors suggested to help eliminate fillers, one should try to create ways to reinforce awareness of the filler word. For example, tapping your own leg when they hear themselves using a filler word. In my view, if you can tap yourself on the leg, you probably have sufficient self-awareness to not use the filler in the first place. More brutal was the idea to have someone else clap when the speaker uses a filler to help recognise the issue. (Zandan, 2018)

I’m delighted that the one process that we routinely guide speakers through to help eliminate or minimise unconscious fillers is ratified by research: simply stop. DO NOT SAY ANYTHING. Most fillers, for example managing a transition between ideas, or reaching for that elusive word, are filling in where that pause should or could be. Until the right word or phrase appears, you have nothing you need to say.

According to Seals and Coppock, speakers may be reluctant to allow nothingness. Initially when eliminating the filler words and using silence instead, the speaker will view the silence as too long. But, critically, silences are interpreted by audiences as having confidence in, and control of, your delivery. Which is the polar opposite of what we worry that silence is saying to our audience. (Seals and Coppock, 2022)

There is also a lot of preparation processes that can fix fillers. People I work with have it drilled into them that there are rhetorical devices which structure your content and are a godsend not just for the audience, but for you too. Mainly because practicing transitions and segues between statements can help reduce cognitive load, which then cuts back on fillers. Summaries, self-responded questions, or even generic transition phrases and being the ‘guide’ to the presentation or speech can be used instead of fillers, and stop you losing track of your place – something like ‘OK, so let’s move on to…’

I get bored of hearing myself say ‘confidence is knowing what comes next’ – but it is true. You can increase confidence, and cut down on the errs and ums, by knowing your communication and plotting your way through it.

Equally, I get somewhat irked when I hear the old chestnut that TED talks require one hour of prep for every minute of talking; that’s a 19-hour commitment. I’m sure you’ve all got countries to run, megadeals to plan, and dinner to make.

Most people prepare really ineffectually, focusing on the wrong things. But, again I am reassured I haven’t been talking BS for the last 20 years when our research showed up two more positive ways to manage the filler issue that we have been advising people weekly.  

First, try and do the practice In front of others, like you mean it (not just a yeah-yeah read through) and more importantly slow down. Practicing in front of others (or at least, using a recording device, your phone or an iPad) increases confidence, and reduces nervousness and fillers.

On speaking speed: I’ve never ever told anyone to speed up. I say ‘slow down’ with astounding regularity. Seals and Coppock reference a study that recorded speakers and asked them if they were speaking too slowly; those who agreed they were, were shown the playback, whereupon they then perceived themselves as speaking too fast

The second thing is about engagement and being present with your audience. Maintaining eye contact has been found to reduce use of fillers, while being reminded to allow hand gestures (see our previous research) shows that we are better able to deliver with clarity, sync with the audience, and reduce filler usage as the speaker better manages cognitive process and is more able to collect their thoughts more fluently.

I said at the beginning that fillers might have a positive. Imagine - if you will - that you have become so reliably fluent and able to communicate through any cognitive load event with consummate ease; like silk slipping off smooth marble. Now is the time to introduce some fillers; add in a thoughtful 'uuhhh', or perhaps a winking 'y'know...' to remind the crowd that you;re human, you're engaged with them, and maybe once upon a time you were, like, uh, kinda, not that sort of, good at uh, public, y'know, speaking.

Guarente + Company is a specialist leadership communications advisor and coaching business. If this blog has piqued your interest, and you’d like to know how we can help, please get in touch with Matt here.

Reference list

Carney, N. 2022. L2 comprehension of filled pauses and fillers in unscripted speech. System. 105, p.102

Conrad, F.G., Broome, J.S., Benkí, J.R., Kreuter, F., Groves, R.M., Vannette, D. and McClain, C. 2012. Interviewer speech and the success of survey invitations. Journal of the Royal Statistical Society: Series A (Statistics in Society). 176(1), pp.191–210.

Duvall, E., Robbins, A., Graham, T. and Divett, S. 2014. Exploring filler words and their impact [Online]. Available from: https://schwa.byu.edu/files/2014/12/F2014-Robbins.pdf.

Oomen, C. and Postma, A. 2001. Effects of Divided Attention on the Production of Filled Pauses and Repetition EBSCOhost. Ebscohost.com. 44(5), pp.997–1004.

Seals, D.R. and Coppock, M.E. 2022. We, um, have, like, a problem: excessive use of fillers in scientific speech. Advances in Physiology Education. 46(4), pp.615–620.

Zandan, N. 2018. How to Stop Saying ‘Um,’ ‘Ah,’ and ‘You Know’. Harvard Business Review. [Online]. Available from: https://hbr.org/2018/08/how-to-stop-saying-um-ah-and-you-know.