Article summary of An experimental test of the relationship between voice intonation and persuasion in the domain of health by Elbert en Dijkstra - Chapter

What is this article about?

This article is about health messages. Paralinguistics can influence the way people react to these messages. Researchers have found that the speed, the pitch, the fluency, the intensity and the intonation of a voice affect how persuasive a message is. This article focuses on the effect of intonation on persuasion.

What is known about intonation?

Intonation can be defined as the variance in pitch while someone is speaking. A few characteristics of intonation are listed below:

• It is used to provide a listener with information about the grammar of a sentence
  • You can hear if a sentence is a question because of intonation
  • People can emphasize certain words (“No, I think HE was the one who did it”).
• It can transfer emotions
• It can transfer certain attitudes

Intonation, persuasion and defensiveness

In research, a relationship between intonation and persuasion has been found. This relation seemed to be linear: as intonation goes up, persuasion goes up as well. This is especially true when there’s high involvement of the listener (so when it’s important to him or her). At the same time, there’s evidence that this relationship has an optimum: when there’s too much intonation, source credibility goes down, together with persuasion. So, there is an optimal amount of intonation.

The authors think that the relation between intonation and persuasion is caused by the emotional information the intonation gives apart from the actual message. When people listen to a health message that is relevant for them, two things are possible:

1. They change their behaviour

According to self-affirmation theory, people want to feel good about themselves. If a health message shows them they are doing something wrong, they feel inadequate. As a result, they change their behaviour to feel better about themselves.

2. They get defensive

A health message can cause feelings of self-threat. Too much intonation may cause an ‘emotional overload’ and people could deal with this by getting defensive. This defensiveness in turn inhibits behavioural change.

The first option is the preferred outcome, because health messages are designed to get people to change their behaviour. So, when do people change their behaviour, and when do they get defensive? The authors think that there’s a difference between people who think that they are healthy and people who think that their health is poor. People with perceived poor health may perceive the message as more relevant and see it as an opportunity to change their behaviour. On the other hand, people with perceived good health have no opportunity to change their behaviour (since they are already doing the right thing) and thus may show a defensive reaction. They used two experiments.

What was the method of the first experiment?

Three health messages about fruit and vegetable intake were used. The research was done using first year students (130 participants) that received SONA credits. They differed in the amount of intonation (low, moderate, high). The researchers had the following hypotheses:

• Moderate intonation leads to more persuasion compared to low intonation in persons who think that they are healthy (1a)
• Moderate intonation leads to more persuasion compared to high intonation in persons who think that they are healthy (1b)
• There’s no difference the amount of intonation in persons who think that they are unhealthy

The message contained two negative outcomes of not consuming fruit/vegetables: increased risk for heart diseases and cancer. It also contained positive outcomes of higher fruit/vegetable intake: looking more healthy, less skin aging, better concentration and lower blood pressure because of vitamins. The voice wasn’t changed by computer, but naturally, by a woman that was trained in voice recording. This improves ecological validity. Messages with high intonation involved more differences in high and low tones compared to messages with low intonation. The messages all sounded natural.

The researchers used a pre-test and a post-test. The pre-test measured perceived own health status on a six point scale (very good...very bad). Intention to start consuming more fruit and vegetables was measured as well as perceived fruit and vegetable consumption, both rated on a five point scale.

The post-test contained a manipulation check that measured perceived voice characteristics. The dependent variable was measured by asking whether the participant wanted to perform a certain health behaviour within one month, six months or five year.

What were the results?

The manipulation check showed that the manipulation worked: participants perceived significant differences between pitch and intonation in the three research conditions. Hypothesis 1a was accepted: moderate intonation is more persuasive in persons with perceived good health than low intonation. Hypotheses 1b was also accepted: moderate intonation is also more persuasive than high intonation in persons with perceived high health.

Hypothesis 2 was not confirmed. The researchers found an unexpected difference between low intonation and moderate intonation. In people who think that they are unhealthy, low intonation is significantly more persuasive than moderate intonation.

What was the method of the second experiment?

In the second experiment, a self-affirmation procedure was used. The authors argue that self-affirmation makes people feel good about themselves. This causes a state of ‘open-mindedness’, which in turn causes people to face threat, instead of denying it. Study 1 showed that for people who perceived their health as good, higher intonation lowered persuasion effects (see Hypothesis 1b). In study 2, the authors try to reverse this effect by using self-affirmation (Hypotheses 2).

The procedure was the same as in Study 1, but now, half of the participants were exposed to self-affirmation. Participants had to select their most valued domain and their least valued domain out of a list of six options (theory, economics, etc). After that, they had to answer several questions. In the affirmation condition, these questions were often related to the most valued domain of the participant.

What were the results of the second experiment?

Hypothesis 2 was confirmed: persuasion in the high intonation condition (and perceived good health) was significantly higher in the self-affirmation condition than in the condition were there was no self-affirmation. In persons with perceived low health, there were no significant differences. They showed high intention in all conditions.

What can be concluded?

This study showed that there’s a difference between the persuasiveness of a health message depending on whether the participant perceived that he/she was healthy or not and voice intonation. Without knowing a lot about the underlying processes, the authors think that this difference is caused by differences in responses to self-threat. People with low health have something to change, so behavioural change is their response to self-threat. People with high health react to threat with defensiveness. This effect can be reversed through self-affirmation.

Because the underlying processes are not clear, the authors cannot say for sure that self-threat is indeed the underlying mechanism. It could also be the case that intonation changes the clarity of a message. Intonation could also have an effect on source credibility, and through that influence (mediate) persuasion.

There are a few limitations to this study:

• Even though participants noticed the difference between moderate and low intonation, the outcomes don’t reflect this. It could be that this manipulation was insufficient.
• The use of a female voice and the fact that this voice was very neutral.
• The use of students.
• The use of an intention measure instead of a behavioural measure. Intentions can differ from actual behaviour.