Curiosity and Motivational States

Going to the conference in Geneva next week, and will talk about curiosity. In preparation, a minor sidepoint about it. The issue is what kind of motivational state curiosity is.

So, first, a partial list of motivational states. They include desires, needs, wishes, hopes, drives, instincts, fears, interests, etc. I don’t mean for the list to be complete, but just to convey some idea of the variety of motivational states. Relevant to my concerns is the different between motivational states with (roughly semantic) content and those without. So, for example, desires can be de dicto, de re, or de se, but all three types are states with content. In the case of de dicto desires, the content is propositional; in the case of de re desires, the content involves (at least) whatever is expressed by a predicative expression that correctly described the content of the desire with respect to the object in question; and in the case of de se desires, the same can be said as with de re desires, except of course that the object in question is oneself (conceived of as oneself).

Other motivational states don’t have content in this way (though they may have content in the way that, say, one’s genetic code has content). Among the content-less are needs, instincts, and interests.

The question is whether curiosity is essentially a contentful motivational state. I’m inclined to think that the answer is “no”, on grounds involving young children and animals: Curiosity is displayed so early in child development, for example, that I don’t see how it needs to involve mental content at all. So, I’m inclined to think of it as a motivational state that can have content but need not.

Thoughts on this most welcome!


Curiosity and Motivational States — 17 Comments

  1. This is an interesting issue. I’m not sure where I’d come down on it exactly, but I’m struck by your appeal to children. Why immediately deny a state has semantic content if children can have it? I think kids have beliefs, desires, and intentions, yet those certainly have content. Do you want to say those states, like curiosity, can have content but need not?

  2. This post made me re-read George Loewenstein’s 1994 article on the psychology of curiosity. In his recap of the history, the old school view was that curiosity was an appetite for knowledge ‘not for any utilitarian end’ (Aristotle) or ‘without the lure of any profit’ (Cicero), and the psychological puzzle about it is why we have the tendency to pursue information even when there is no obvious extrinsic reward in it. (I take it Aristotle and Cicero are right that we don’t count it as curiosity when information-seeking behavior has an obvious expected utility benefit.)

    At least the way Loewenstein tells the more recent story, there was a period in the first half of the 20th century when curiosity was regarded as a primitive drive, generating an aversive state relieved only by information-gathering behavior (Freud, behaviorists like Thorndike). But he doesn’t cite any psychologist who still maintains that sort of view now, and one reason he mentions is that the drive view has trouble making sense of why we voluntarily seek out curiosity-inducing situations.

    I think the dominant views now would all characterize curiosity as a contentful state, either involving some kind of desire for making sense of the world (Hebb, Piaget, Hunt), or a desire for cognitive competence (White), or — this is Loewenstein’s own view — a desire to mend a gap between one’s actual and one’s desired information level.

    But I have to say curiosity is still not terrifically well understood, and the existing scales for measuring it are all a bit dismal. If you have a bit too much time on your hands, google up something like the State Epistemic Curiosity Scale, you’ll see what I mean.

  3. Very interesting issue. For starters, how about this as an argument that curiosity has content:

    1. Being curious requires that you conceive of what you are curious about. (So for example, being curious about the grand canyon requires have some concept the content of which is the grand canyon, being curious about physics requires having some concept the content of which is physics, and so on).

    2. Therefore curiosity has content.

    Leaving that aside, here’s a shameless plug: I’ve got a paper on curiosity forthcoming in PPR which engages these issues. It gives a theory of what curiosity is, engages previous work on the subject including Jon’s, and address both motivational issues and semantic issues.

    Fred Schmnitt and Reza Lahroodi have a recent paper on all this too…

  4. Jennifer, useful info, thanks, I’ll take a look at the curiosity scale. Josh, no, I don’t think the same about the other states you mention, one of which (desires) I explicitly addressed. The reason young children are interesting here is that they seem to be curious too early to have much conceptual apparatus, but I’m doing too much armchair work here to be confident. Dennis, I think the first premise of your argument isn’t compelling. Kids try to taste everything very early in life, but they don’t have the concept of taste–they exhibit curiosity about whether things are edible in spite of that. I exhibit curiosity by looking to my left when there is movement there in my visual field, but I’d do that even if I didn’t have the concept “to my left”, as I did when I was an infant and did the same.

    I think your examples show at most that sometimes curiosity involves conceptual elements. So maybe it is a motivational state that may or may not have content. Or maybe it is a motivational state that sometimes triggers a desire, and desires always have some content, either propositional or non-propositional.

  5. Interesting points, Jon, but I don’t think I’m convinced. Kids may well try to taste things before having the concept TASTE. In such cases, they may well curious about objects such as grass blades, wondering e.g. whether those grass blades are edible. But what they must do to be curious about a given grass blade is: conceive of that grass blade. Grass blades are what these kids are curious about, and so grass blades – not taste – are what they have to conceive of, in order for them to be curious about grass blades. Or at least, that is what premise (1) in the post above is intended to claim. And that claim seems true, despite your case about the concept of taste.

    Similarly with the case of looking to the left. Presumably, when you see some movement to your left, you get curious about the area to the left of you. But this does not, either intuitively or by premise (1) above, require you to possess the LEFT OF concept. You just need some concept of the area to the left of you. Perhaps, in looking to your left, you are wondering what is over there, or wondering what was just moving, or wondering whether that thing you just saw is edible.

    Each of these ways of wondering (if you like to boxological talk: ways of having a question in your curiosity box) is a way to be curious about what is to the left of you, without having the concept LEFT OF. And so, in this case as well as your other, we do not see a person being curious about something while not conceiving of that thing. As a result, these cases don’t give us reason to reject premise (1). No?

  6. Dennis, I think some of the experiments about mental and submental states would be useful here. When behavior engages only submental states, there’s a time difference in the response to a stimulus from the response time when mental states are involved. That’s why looking in the direction of motion detection is an interesting example. One would have to do the experiment to find out, but the response is so fast, I expect the results would fall in line with other cases where the response doesn’t run through mental states. And that would fit with my own description of the case: motion detection is followed by looking, direct and immediate. There is no intermediate thing such as wondering what’s over there.

    I also think you are thinking of much older children than I am. Think about children that are, say, 3 months old. They respond to new sensory stimuli, reaching and grabbing in the process. They of course have no concept of what is passing across their visual field, and once again, maybe there is conceptual content here that is engaging and plays a role in the resulting behavior. But there are empirical tests for this sort of thing, and the time difference between stimulus and response is one way to test whether the response involves mental states.

  7. Jennifer, a quick follow-up about this remark of yours: “(I take it Aristotle and Cicero are right that we don’t count it as curiosity when information-seeking behavior has an obvious expected utility benefit.)”

    I wonder what you think about situations like being curious about how the stock market works: whether, for example, the random walk theory is likely to be true. I am curious about such things. If you ask me why, I will at least use an epistemic modal: it might allow me to invest more wisely. So I am curious, but at least open to connections to utility.

    Such a case is not one where there is an “obvious expected utility benefit,” but there is a difference, I would think, between explaining our behavior in terms of our curiosity and explaining it just in terms of curiosity. I used to teach probability games to logic students that were classic scam games. Attendance went up! Why? If students said, “I’m just curious,” they were probably lying. But they might have been curious as well as motivated by practical interests. No?

  8. Hi Jon, yes, you are right, one’s search for information might be overdetermined. Aristotle and Cicero might still be right that we don’t ordinarily *count* it as curiosity when there is obvious benefit for the subject in finding something out — just thanks to our general (lazy!) psychological preference for single-cause over multiple cause explanations. If I’m checking the thickness of the ice as I’m about to start skating on it, typical ascribers probably wouldn’t count that as curiosity on my part, even if I turn out to be one of those people who is just fascinated by ice thickness issues whether or not they present the prospect of harm or benefit. But the psychological availability of the easier explanation shouldn’t overshadow the real appropriateness of the more complex one.

    If one’s search for information on the random walk theory were motivated strictly by the desire to invest more wisely, that search would not manifest curiosity (do you agree?), but yes, the search for information might be motivated both by curiosity and by desire for gain.

    I’m a little surprised by your suggestion that ocular response to motion detection is going to count as manifesting curiosity. I would have thought curiosity needs to be more purposive. On your view, is it curiosity when pupils dilate in response to lowered illumination? Just curious about that.

  9. I think I’m more of a pragmatist here. Suppose you think that everything we do is directed toward happiness or flourishing or some such thing. Then everything, including what we are curious about, has some extrinsic motivation of some sort. So I’m suspicious of unqualified claims about intrinsic motivation. I think they can be qualified to avoid my worries, but I’m not sold in the idea that curiosity is logically independent of other interests. To hold that view, one has to explain away the obvious acceptability of saying things like this: “I am curious why my left ankle hurts when I twist it this way.”

    I think pupils dilating is at the reflex end of the spectrum. Blinking is halfway between: it can be under our control, but typically isn’t. Attention is closer yet to things that we typically control, since we can avoid looking and do so forever (though we can’t with blinking). Maybe infants can’t though; but even if they can’t, it is of a kind over which they’ll develop such an ability, and the activity in question will be motivated the same way once such an ability is in place, or so I would think. The central question is how to distinguish attention and focus when they don’t involve curiosity from those that may. I don’t know how to do that, but I’m suspicious of the idea that curiosity never overlaps with attention or focus of the sort that doesn’t involve any passional or emotional intensity. That seems to me to be a mistake on Lowenstein’s part, one arising from a typical philosopher’s mistake of focusing on highly sophisticated intellectual examples of the phenomenon in question. Curiosity about the completeness of arithmetic might involve features quite different from infant curiosity, and philosophers have a penchant for using examples that are much too intellectual and then generalizing like crazy from the small and unrepresentative sample. So, though I’m not positive the peripheral vision example is an example of curiosity, I’m inclined that way, and think that arguments that derive from the history of thought about curiosity are susceptible to the unrepresentative sample problem just described.

  10. Jennifer, I like your ice example. I wonder whether I am idiosyncratic in my speech patterns, so I’ve checked with some others here about what to say in that case. So suppose you are checking the ice thickness, and I ask, “Why is she so curious about the thickness of the ice?” That seems like a perfectly sensible question to ask, doesn’t it? And the answer would have to appeal to something about your practical interests.

    What is true is that, in the case in question, you wouldn’t be investigating just out of curiosity. Being just curious is a more restrictive notion, it seems to me, than being curious.

    If we change examples, and we find a student taking business classes and studying for an exam, if I ask, “why is s/he so curious about international finance?”, now the appropriate reply seems to be, “well, as far as I can tell, s/he’s not curious about it, s/he just wants a high-paying job after graduation.” But in the ice example, the same kind of reply doesn’t seem appropriate to me: “she’s not curious, she just doesn’t want to fall through the ice.” Does it seem that way to you too, or not?

  11. Jon, I like your intuitions, although I wonder if there are some pragmatic effects here. I think studying for an exam in international finance is the kind of very recognizably goal-driven behavior that is going to make an ascription of curiosity conspicuously inappropriate, in a way that will invite rebuke.

    I would love to think that sometimes my students cramming in the hallways outside the final exam are having moments of curiosity, but frankly, at that point, I doubt it is psychologically possible. I fear the right answer to the question, “Why are those students so curious about the Fourth Meditation?” is “hahaha”.

    If you were to encounter someone who was searching for information in a somewhat more peculiar way, say, eavesdropping in the financial district, then because this behavior is less recognizably motivated by the search for extrinsic gain, it would sound OK to ask “why is she so curious about what those suits are saying?” even if the answer were the same: “because she is collecting information that will help her get a high-paying job after graduation”.

    But I admit that in this case, as in the ice case, it is possible to be motivated both by curiosity and by desire for gain.

    On the ice case, if it were tremendously clear to you and your interlocutor that I was about to go skating, I think you wouldn’t naturally describe me as being curious about the thickness of the ice. I think that question only makes sense when it’s unclear what my gain would be in knowing. And then I think there might be a pragmatic explanation of why we don’t feel right about answering “She’s not curious, she just wants to check whether it’s safe to skate” — that would carry the implication that you’re clueless for having failed to recognize my intentions, which is probably a bit mean in the skating case. (In the case where someone asks about the studying student, the fact that studying has so obvious an end makes it more acceptable to reprimand the presupposition of the question.)

    On your other point, I have to defend George Loewenstein from the charge of making “a typical philosopher’s mistake of focusing on highly sophisticated intellectual examples of the phenomenon in question”. He’s all over the low-level examples, too, and in his experimental work has focused quite a bit on puzzles where subjects click on little bits of pictures to bring them into focus (interesting fact: we get much more excited about visual information when we are on the verge of seeing the whole picture).

    You are right that he maintains that curiosity requires “emotional or passional intensity”, however. Some psychologists (Buhler, in her classic work on the first year of life) classify all moments of turning towards strange sights or sounds as curiosity; Loewenstein claims that “such orienting reflexes have more in common with the modern term ‘attention’ than with curiosity as defined in the pre-modern period. They are not necessarily intrinsically motivated, are unemotional in character, and lack the drive properties associated with a cognitive appetite.”

    Loewenstein would absolutely agree that what motivates people to examine the completeness of arithmetic is quite different from what motivates infants to explore their environment (although he notes Berg & Sternberg’s work correlating infant response to novelty and later IQ, which might come in handy on the arithmetical questions, assuming adequate adult curiosity on them). In Loewenstein’s view, if we try to posit a single notion of curiosity to explain both of these things, it’s going to be a uselessly weak notion.

  12. Jennifer, I agree it might be pragmatic factors at work in the examples. And also that Lowenstein doesn’t focus only on high-level examples–I was actually thinking about the argument you quoted, which I think isn’t a very good one. I also think having a single notion covering all these instances needn’t be excessively weak. For example, it might be just the notion needed to explain why true beliefs are valued for their own sake. I’m in Geneva, and Michael Brady’s talk yesterday was about such an attempt, and though he was pessimistic about it, it was on grounds that not all true beliefs are valued in this way. Apart from that issue, the reliance on this general notion of curiosity presents a really nice story about the final value of true beliefs: they are valued for their own sake, on the basis of something extrinsic to them. So whether the notion is too weak depends, I think, on the theoretical context one is envisioning.

  13. Like Brady, I’m pessimistic about using curiosity to explain why true beliefs are valuable.

    Consider a true belief, and a false belief, neither of which has anything to do with what one is curious about, and across which other things (such as justification etc) are also equal. For instance, compare two very similar cases, both where you come to hold a belief by listening to the radio. In the first case, you come to believe that the number of grass blades on your lawn is even. In the second, you come to believe that the same number is odd. (Perhaps the radio hosts are fooling you into thinking they had an army come count this morning.)

    In one of these cases your belief is true and in the other it is false. Neither of these beliefs satisfies or is in any other nontrivial way related to your curiosities. Nonetheless, the true one is better. Indeed, it has more final value. Hence, the final value of true belief cannot be fully explained via any relationship between true belief and curiosity.

    [You might object that the true belief here actually does not have any more value than the false belief. But it does, despite its being utterly trivial. One way to see why is to consider (a) believing all the true trivialities, and (b) believing all the false trivialities. Surely (a) is better than (b). But then, each of the trivial true beliefs must make some positive evaluative difference, so that their values combine to make (a) better than (b). (It isn’t like (a) is better than (b) in some kind of organic unity way). And so, since having all the trivial true beliefs is better than having all the trivial false beliefs, having any trivial true belief is better than having any trivial false belief. And so, the trivial true belief I was talking about really is better than the trivial true belief I was talking about above.]

    This is not to deny that the final value of true belief is *partly* explained by its relationship to curiosity. (In my view, fwiw, the final value of true belief *is* in that way partly explained). Curiosity just can’t *fully* explain the final value of true belief.

  14. The issue –it seems to me–revolves then
    around what mental content consists in and
    what shows mental content; decide what the signs are of mental content generally and see if curiosity shows the signs.
    But surely there could be curiosity sans
    mental content– a curious cat perhaps–and curiosity with content-Sherlock Holmes.

  15. Won’t de re thought be able to get you a whole lot of young kid thoughts? “What’s that?” or “What’s This?” don’t seem too far from what I’d be inclined to attribute to a child reaching for something.

  16. Hi Mark, demonstratives can do some work here. I still think, though, that time-lapse experiments would be useful (as well, perhaps, as some neuroscience on brain areas at work in expressions of curiosity–to this point, it seems that the location is in the same area in which deficiencies of activity seem involved in autism; what to make of that is not clear, though I’d be hesitant to explain non-autistic behavior in terms of conceptual content).

  17. A late and small point on the Whitcomb-Kvanvig dispute re the mental nature of curiosity.

    I sometimes find myself lingering in certain places (e.g., coral reefs, bars). Often it seems right to say that my lingering is explained by my curiosity. But it seems wrong to say I have conceived of anything that I am curios about. (I’m ruling out, perhaps unfairly to Whitcomb, very generic ascriptions such as being curious about that which will happen next). So, it seems to me that curiosity may be non-conceptual. (It is much less clear whether there are lingering cases in which it is wrong to say, de re, that there is something of which I am curious).

    What I want to tentatively suggest, however, in the cases of lingering, the cause of the lingering (i.e., the curiousness) is identical with or associated with a certain phenomenal state. Indeed, it is hard to think of a convincing case of curiosity without a phenomenal component. So, it may be that the distinctively mental aspect of curiosity is phenomenal rather than representational. (Of course, Kvanvig will have to argue contra theorists who attempt to reduce the former to the latter).

    Note also that the object of the curiosity is sometimes an inconceivable phenomenal state. Foodies, virgins and aspiring bungee-jumpers know what I’m talking about.

Leave a Reply

Your email address will not be published. Required fields are marked *