Theory of Mind

What is Distinctive about Theory of Mind Explanations?

We can understand a good deal about everyday behavior just by thinking about observable actions. We see Gill with an empty glass looking around the room and we suppose that she will locate the drinks table and approach it. We see her reject the red wine and take the white wine, and we can predict that she will do the same in the future. But thinking about Gill's beliefs and desires gives additional purchase on the problem of explaining and predicting her behavior. Gill's behavioral history of seeking white wine would not enable us to understand why she initially approached the red wine on this occasion. But if we infer mental states—by supposing that she likes white wine and mistakenly thought this was what was in the first bottle—then we can make sense of her initial selection, her disgust on finding out that she was wrong, and her reselection of another bottle. Such theory of mind explanations gives a more complete account of Gill's actions than explanations that make reference only to observed behavior. However, this comes at the cost of complexity and uncertainty in formulating such explanations.

The key problem is that beliefs, desires, and intentions cannot be directly observed, and must instead be inferred on the basis of observable evidence and background knowledge. Against [Page 1000]everyday intuitions, even perceptual states such as “seeing” cannot be observed in a straightforward manner. To illustrate, imagine that you observe a reader and a nonreader viewing a piece of text. Both people show the same observable signs that they are “seeing”—in that both are oriented toward the text and appear to be paying attention to it. However, these observable signs give no notion that the reader sees meaningful words on the page, whereas the nonreader sees meaningless squiggles. Even the apparently simple case of seeing requires us to make inferences that go beyond the observable data in order to understand what might actually be seen. And of course this rapidly increases in complexity if we wish to take account of the many interactions between beliefs, desires, perceptions, and intentions, as we might when evaluating the guilt of a defendant in a court case or the intentions of a lover who is not returning our call. These considerations make theory of mind seem more like reasoning than perception, and this is indeed the prevailing assumption behind many studies of theory of mind. But as we shall see, there are good reasons for supposing that at least some aspects of theory of mind need to be more like perceptual processes than is often supposed.

...