Most people can tell the difference in the faces of those who are jubilant over good news and those who are distraught over bad news. But the same does not hold for voices of people whose faces are absent.
Humans cannot distinguish between screams of joy and cries of terror, according to a new international study published in the Journal of Experimental Psychology:General.
Conducted by doctoral student Doron Atias and led by Prof. Hillel Aviezer of the Hebrew University of Jerusalem’s psychology department, it was conducted in cooperation with Prof. Alexander Todorov of Princeton University and colleagues from Mifal Hapayis, Israel’s official lottery. They found that voice expressions do not contribute to the distinction between extreme emotions. Our intuitions are thus called into question.
Vocalizations were rated by 39 Hebrew University students who listened to 58 vocalizations (29 positive, 29 negative). During the course of the study, the researchers asked the participants to hear voices from a variety of exciting situations from everyday life. The participants failed to recognize emotions based solely on vocal expression.
“Many technology companies are currently engaged in the development of computerized means to identify emotion expressions in a voice, but it is possible that the distinction between these expressions in everyday life does not allow for good identification without information about the context,” noted Aviezer.
In one of the experiments, American subjects from Princeton University in New Jersey, US, were asked to listen to recordings by Israelis who had won first prize in the lottery. Many Israelis are well familiar with the fact that a woman known by her first name as Erela of the National Lottery calls winners to tell them that they have won a major cash prize. The winners’ responses are recorded and saved and sometimes presented in TV commercials for Mifal Hapayis.
The unique recordings of winners of the Israeli national lottery made it possible for the researchers to examine how changes in the intensity of the emotional experience (for example, in response to low vs. high gain) affect the quality of the vocal responses in these situations. The vocal responses of the winners were heard by the American respondents, who were asked to assess how positive or negative they were, but without knowing their source or understanding the Hebrew language.
The results of the experiment showed that while the responses of the winners of the smaller amounts (up to 125,000 New Israeli Shekels) were perceived as positive by the listeners, the responses to the high prize amounts (250,000 New Israeli Shekels or more) were perceived as negative. These results reinforce the claim that the distinction between positive and negative expressions is obscured in extreme situations and show that the intensity of the emotional experience plays a role in this ambiguity. It seems that such vocalizations do not express inner emotion, but are generated as a tool for influencing others in a flexible context-dependent manner.
Other experiments reinforced this conclusion of the researchers. American subjects were greeted with authentic cries of joy from parents who unexpectedly met their soldier son who returned home after a long time on the battlefield. On the other hand, the audience heard authentic screams of fear from people during a confrontation with a threatening burglar who broke into their home. The subjects failed to distinguish between the positive and negative expressions they heard, when in fact they judged all the responses as negative.
This piece of research, titled “Loud and Unclear: Intense Real-Life Vocalizations During Affective Situations Are Perceptually Ambiguous and Contextually Malleable,” follows a similar study carried out six years ago by Aviezer in cooperation with Prof. Jacob Tropeh of New York University and Prof. Todorov of Princeton. They examined responses to facial expressions of tennis players and found that they could not discern a difference between those who won and those who lost.
“This study’s findings shed new light on all that we knew about vocal expressions of emotion and contradict the prevailing view of the general public and researchers that the overlap between the positive and negative extremes is not possible.”
The research shows that at least at the peak of emotion, the distinction between negative and positive expressions is blurred. “In my opinion,” concluded Aviezer, “many technology companies that are currently engaged in the development of computerized means to identify emotive expressions through voice expressions will not succeed in their mission, because the distinction between these expressions in everyday life does not allow for accurate identification without information about the context.”