Speed of Speech and Persuasion: Why Faster Speakers Sound More Credible

By Noam Shemla · The Studies

TL;DR Miller and colleagues played the same recorded argument to strangers in parks, malls and streets around Los Angeles, spoken either fast or slowly. The fast version was more persuasive, and the speaker was rated as more knowledgeable. The authors tested three explanations; only one fitted: speaking quickly works as a cue that the speaker knows what they are talking about.

• Paper: Miller, N., Maruyama, G., Beaber, R. J., & Valone, K. (1976). Speed of speech and persuasion. Journal of Personality and Social Psychology, 34(4), 615–624.
• Design: Two field experiments disguised as street interviews for a radio programme. Listeners heard a taped opinion, then rated how far they agreed with it and what they thought of the speaker.
• Participants: 359 people approached in public places around Los Angeles (Experiment 1) and 90 shoppers at an open-air mall (Experiment 2).
• Varied: Speaking rate (slow or fast; then slow, moderate or fast), crossed with the speaker’s credibility (Experiment 1) or the complexity of the message (Experiment 2).
• Measures: Agreement with the message and ratings of the speaker, all on 10-point scales.

“Fast talker” is not a compliment: the phrase itself is a warning. Norman Miller and colleagues at the University of Southern California asked whether speaking fast actually persuades, and if so, why.1

Rate varies a lot between ordinary speakers. Speech textbooks of the time put normal speech between 120 and 180 words a minute, so the question was not about freakish speed but about where in that everyday range a speaker sits.

Three reasons speed might matter

Each explanation the authors considered makes a different prediction, which is what let two experiments tell them apart.

Three explanations of how speaking rate could affect persuasion

An earlier pair of laboratory experiments by two of the authors had already found fast speakers more persuasive and more trusted. But in a laboratory, participants may listen as if they will be tested, which could hide how speed normally affects understanding and arguing back. So these experiments moved to the street.

The street interviews

An interviewer, one of 14 undergraduate experimenters, stopped people in parks, shopping malls and neighbourhoods, introduced themselves as working for a radio programme, and asked them to listen to a listener’s taped opinion and react to it.

The two field experiments

In the first experiment, both manipulations pushed in the expected direction. The biochemist was seen as far more knowledgeable than the locksmith, and the fast speaker as somewhat more knowledgeable than the slow one. Both the fast and the credible versions persuaded more.

A deliberately strict test

Fourteen inexperienced experimenters ran the first study, and their results differed from one another. The authors could have treated them as interchangeable, as most studies treat their single experimenter. Instead they tested each effect against how much it varied between experimenters, so that a significant result would hold for experimenters in general, not only these fourteen.

That choice cost them significance: with only 13 degrees of freedom in the denominator, the effects of speed and credibility on agreement fell just short. The authors also found a stutter over one word in the fast, credible recording, the kind of slip earlier research had shown lowers a speaker’s credibility. Taken together, they read the first experiment as consistent with the credibility explanation and against the other two, rather than as proof of it.

A topic nobody could argue with

The second experiment was designed to give the counterargument explanation its best chance. Hydroponics was chosen because almost nobody knew what it was, so listeners had no ready-made objections to fall back on and would have to process the arguments themselves. If fast speech worked by squeezing out time to argue back, it should have mattered most for the simple version of the message, and hardly at all for the complex one, which would absorb a listener’s full attention at any speed.

It did not: complexity made no difference, and there was no hint of the predicted interaction. Speed did.

The authors concluded that speaking rate works mainly as a cue about the speaker rather than by changing how the message is processed, and drew a broader lesson from it: listeners often judge a persuasive message by quick impressions of who is delivering it, not by careful scrutiny of what is said. Their suggested new rule was “Beware of the fast talker”.

They were careful about what they had not shown. A speaker asked to talk fast may change other things too, and only computer-altered recordings could isolate speed itself. Comprehension was never measured. And in a real conversation, where a fast talker leaves no pauses for the other person, speed might frustrate the listener instead.

Has it held up?

The first half has. When recordings of six real voices were sped up and slowed down by computer, by up to half, listeners rated the faster versions as more competent, steadily and in a straight line; they rated them as most benevolent, however, at their natural rate.2

The second half, that fast speech persuades whatever the circumstances, has not held up so simply. Later experiments with involving topics found that speed depends on what the listener would otherwise do with the message:

What later experiments added

So speed does lend a speaker authority, but it also makes a message harder to think about, and whether that helps depends on what thinking about it would have done.34 Miller and colleagues were right that the credibility cue is real; the rule they joked about, that fast talkers always win, was too simple.

What this study cannot tell you

• It cannot separate speed from everything that comes with it. One speaker per experiment recorded every version, and changing speed may change pitch, rhythm and energy too. The authors say so themselves.
• Its effects are modest and not all significant. In the first experiment the effects on agreement fell short of significance under the strict test the authors chose; in the second, two of the four speaker ratings only approached it.
• Comprehension was never measured, so the conclusion that fast speech did not hurt understanding is an inference from the results, not an observation.
• These were one-way recordings, heard once, about topics the listeners had little stake in. Later research shows that stake matters.
• Small cells. The second experiment had 90 people spread over six conditions, about 15 each.

What a speaker can take from it

The practical lesson is less “talk fast” than “don’t be slow”. In the second experiment, agreement barely moved between a moderate 140 words a minute and a fast 191; the damage was done by the slow 111. A slow, halting delivery reads as a speaker unsure of the material, and listeners mark the message down with it.

Pace is also something listeners use when they are not scrutinising the argument. When your audience is sceptical, a brisk, fluent delivery lends you authority. When they are on your side and you want them to think it through, a pace that leaves them room to reflect serves you better. The skill is not a fixed speed but choosing one.

This article summarises a published paper, read from a scan of the journal article. Every mean, rate, sample size and test statistic is the authors’ own, from the results of both experiments and the footnotes. The later findings are summarised from the published abstracts.

Measured

• Agreement with a recorded message, on a 10-point scale
• Ratings of the speaker’s knowledge, intelligence and objectivity
• The speaking rate of each recording, in words a minute

Inferred, not measured

• That speed works as a credibility cue: inferred from which predicted patterns appeared and which did not
• That speed itself, rather than something that comes with it, caused the effects: the authors say it could not be isolated

References

1. Miller, N., Maruyama, G., Beaber, R. J., & Valone, K. (1976). Speed of speech and persuasion. Journal of Personality and Social Psychology, 34(4), 615–624. https://doi.org/10.1037/0022-3514.34.4.615
2. Smith, B. L., Brown, B. L., Strong, W. J., & Rencher, A. C. (1975). Effects of speech rate on personality perception. Language and Speech, 18(2), 145–152. https://doi.org/10.1177/002383097501800203
3. Smith, S. M., & Shaffer, D. R. (1991). Celerity and cajolery: rapid speech may promote or inhibit persuasion through its impact on message elaboration. Personality and Social Psychology Bulletin, 17(6), 663–669. https://doi.org/10.1177/0146167291176009
4. Smith, S. M., & Shaffer, D. R. (1995). Speed of speech and persuasion: evidence for multiple effects. Personality and Social Psychology Bulletin, 21(10), 1051–1060. https://doi.org/10.1177/01461672952110006