Based on excerpts from this article.
What do facial expressions tell us? What kinds of information can we gather from seeing the expressions of others? Whether or not we are given contextual clues, what kinds of things can we presume about another person’s state based on their face? Should we consider these as messages sent to us, a form of communication via facial expression, or are they involuntary expressions of an internal state?
For example, consider the expression shown by the woman in this photograph I took in 1967 when I was in the highlands of what is now called Papua New Guinea. Consider the diverse information that someone might obtain when observing this expression, totally out of context, just as it appears here:
- Someone insulted/offended/provoked her.
- She is planning to attack that person.
- She is remembering the last time someone insulted her.
- She is feeling very tense.
- She is boiling.
- She is about to hit someone.
- She wants the person who provoked her to stop what he/she is doing.
- She is angry.
Compare this to the information that could be obtained from the expression shown by another person from Papua New Guinea:
- Someone or something revolted him.
- He is thinking about how to get rid of it.
- He is remembering the last time he was revolted.
- He is feeling nauseous.
- He feels like he’s on a roller-coaster.
- He is going to leave.
- He wants the person who revolted him to stop what they are doing.
- He is disgusted.
What do facial expressions tell us?
Each facial expression of emotion communicates very different information, yet they all potentially provide information about the same seven kinds or domains of information, including:
- the antecedents, the events that brought about the expression
- the person’s thoughts, plans, expectations, memories
- the internal physical state of the person showing the expression
- a metaphor
- what the expresser is likely to do next
- what the expresser wants the perceiver to do
- an emotion word
We do not know which information domains those actually engaged in a conversation derive from each other’s expressions. It could be only one information domain or all of them. Which it is may depend on who those people are, how well they know each other, what they are talking about, their social class, and their culture. In other words, we do not know the answer, and there is no certain way to find out.
The people involved in the conversations could not tell us. Even if we interrupted an individual and asked them what they thought when they saw a particular expression on the other’s face, we would not necessarily find out. For they could only tell us what they think happened. Although that is interesting to know, it is likely to be a retrospective construction, not what actually happens when the expression first registers. The initial translation of an expression into some meaning (any one of the information domains listed) is likely to be so immediate that we are not aware of the process we go through. Darwin noted this in saying, “it has often struck me as a curious fact that so many shades of expression are instantly recognized without any conscious process or analysis on our parts” (1872/1955, p. 359).
There are exceptions, of course, when we ponder the meaning or significance of an expression. This happens when an expression is unusual or its occurrence at a particular moment in time is incongruous with everything else that is happening. Then the persons trying to figure out the meaning of the expression are quite aware of their thought processes, but these may not be the thought processes that are typically involved when expressions are translated into meaning immediately.
Although we cannot find out exactly what people actually do, we can find out what they can do. We can determine if each of the domains of information listed earlier can be derived from an expression. We know that if we show people a facial expression of emotion they will agree in their choice of which emotion word (anger, fear, etc.) fits the expression. But what if we asked them to choose which plan a person is making, which event preceded the expression, which sensation might be felt, and so forth? The people I studied in Papua New Guinea had no trouble identifying the events associated with that particular facial expression. Rosenberg and Ekman 1995 also found that Americans show similar agreement. To my knowledge no one has yet determined whether each of the information domains I listed can be derived from facial expressions. That work could be done, and I expect it would find that agreement is good for each domain of information. Facial expression in our communications can provide, I believe, each of these different types of information, but that is not a demonstrated fact within or across cultures.
I think we use emotion words (anger, fear, disgust, sadness and so forth) as a shorthand, an abbreviated way to refer to the various events and processes that make up the phenomenon of emotion. Each word refers to a different set of these organized, integrated processes. When someone says or thinks “that woman is angry or that man is disgusted,” we do not know which of these events are processes they are considering or if they are actually considering any of them. It is much more convenient, if less precise, to use the single emotion term and to list, as I have in these examples, the various information domains that term encompasses. But remember, as my examples show, I know very different sets of information for anger than for disgust, and so on.
Test Your Knowledge of Facial Expressions
How well can you decipher facial expressions? Find out by taking our facial expression quiz.