Are epistemic emotions metacognitive?

Abstract

This article addresses the question whether epistemic emotions (such as surprise, curiosity, uncertainty, and feelings of knowing) are in any sense inherently metacognitive. The paper begins with some critical discussion of a recent suggestion made by Joelle Proust, that these emotions might be implicitly or procedurally metacognitive. It then explores the theoretical resources that are needed to explain how such emotions arise and do their work. While there is a perennial temptation to think that epistemic emotions are somehow about the cognitive states of the person undergoing the emotion, we will see that such views can and should be resisted.

Keywords:

• Implicit
• memory
• metacognition
• procedural
• surprise
• uncertainty

Acknowledgments

I am grateful to Joelle Proust and two anonymous referees for their comments on a previous draft of this paper.

Notes

1. This is even true, I think, of what psychologists call processing fluency. While many of the effects of fluency are known to be mediated by naïve metacognitive theories (Alter & Oppenheimer, 2009), others are more direct, caused by the positive valence that is produced by fluent processing (Topolinski & Strack, 2009). Although processing fluency is treated at some length in Proust (2014), it will not be discussed here. This is because it is not normally regarded as an epistemic emotion. Nor is the notion familiar to common sense in the way that epistemic emotions like surprise and uncertainty obviously are.

2. Note that the “double accumulator” account that Proust (2012) uses as her central example of a procedurally metacognitive system is a predictive model not unlike those described here. I think she can only be tempted to believe that the system is a metacognitive one because it is used to explain uncertainty-monitoring behavior, which is itself intuitively metacognitive in nature, as we will see. Once control systems of the “double accumulator” sort are seen to be well neigh ubiquitous in cognition, all temptation to see them as metacognitive should fall away.

3. Nagel (2014) uses a very different example to make essentially the same point as the one made in this section. She points out that cross-modal sensory integration appears to qualify as metacognitive by Proust’s (2014) criteria. But no one seriously thinks that it is.

4. Attention is not explicitly metacognitive, of course. For it doesn’t represent experiences and memories as such when deciding how to direct attentional signals, nor is there an explicit intention to render the resulting states conscious. Rather, the intentions take the form, switch [attention] to that, where the indexical picks out the first-order content represented by the percept or memory in question.

5. Recall, however, that information-seeking is not really metarepresentational in nature; it is motivated, rather, by a first-order question-like state.

6. Indeed, a number of researchers have noted the commonalities – and likely partially shared mechanisms – involved in both spatial search and memory search; see Hills (2006); Hills and Dukas (2012).