Instrumentality Boosts Gratitude

5 Comments

• Hugo Mercier 14 March 2011 (13:47)

  Thanks to the two of you for this great paper. A couple of questions/comments. As many emotions, gratitude probably has a double role: one of motivating behavior (in this case, reciprocating behavior), and one of communication (in this case, communicating indebtedness I guess). It's possible that there's less need to communicate indebtedness after the help has been received because both helper and receiver are mostly aware that there has been costs incurred, etc. It's still good to show we understand we indebted, as a receiver, but the helper seems to already have a pretty good idea of how we feel (as in the results of your exp. 2). By contrast, before the help, the helper may not know exactly what the receiver wants. More specifically maybe, she may not know how much the receiver would be willing to be indebted if she received the help. So feeling (and showing) more gratitude before help has been given may make sense from a communication point of view. (Admittedly, in the first experiment there can be no communication between receiver and helper, but that is a rare situation for which our emotions are presumably not adapted to quote.) Also, I would maybe hesitate to apply the advice you give helpers at the end (to ask for favors while they are still helping). This may capitalize on the receiver's increased feelings of gratitude, but isn't there a chance that the receiver may then perceive the request as a form of blackmail (or at least a market transaction)? It seems to me that the exchanges in which we feel gratitude are usually outside of 'market' transactions in which the things being exchanged are clear from the start (you usually don't feel much gratitude towards the office supply salesman). By asking for a favor before, or as the help is being given, the helper may extinguish any feeling of gratitude altogether, by switching to another category of exchange.

• Ben Converse 14 March 2011 (21:38)

  Thanks, Hugo, for the interesting thoughts. And, even more so, thanks for convening this conference! As so often happens, I’ll take the second question first. The good news about it being a web conference is that I won’t then finish my answer and say that I forgot what the first question was. You were reluctant to accept (the all-too-short form of) our strategic social exchange advice, and based on the version of it that you’re thinking about, I’d say that’s a wise reluctance. I’m really glad you brought this up because it will allow me to expand on what we’re thinking here. The potential effectiveness of this strategy would depend so much on how the helper wields it. The helper who makes his reciprocation expectations clear to the beneficiary is probably setting himself up for the negative outcome that you’re describing. Whether through an explicit request or a poorly-dropped hint, if the helper’s ulterior motives are obvious, it’s likely to lead to the kinds of “blackmail” feelings that you suggest. This would be a lot like the situation described by Ames, Flynn, and Weber (2004) where beneficiaries are less impressed when they think helpers decided to help on the basis of cost-benefit calculations. Our advice, then, depends on the well-dropped hint, or subtly extracted commitment – that is, it depends on getting the beneficiary to agree without recognizing that he’s making a trade. Of course, in some other contexts it may be perfectly appropriate to do it more explicitly. In many negotiations, for example, it would be considered perfectly acceptable to make explicit trades. You also noted in this comment that gratitude simply may not be relevant to some exchanges. I agree, but only in part. You’re right on the mark that the research shows gratitude will be greater for gratuitous benefits, beyond what is expected by role norms. We probably aren’t particularly grateful to the bagger who packs up our groceries for us. Later in the parking lot, when the bag rips and food spills everywhere, we would feel gratitude for the stranger that stops to help out. So, same action – putting food in the bag – but very different levels of gratitude. That said, I would argue that we can and often do still feel gratitude in many, or at least some, market-oriented exchanges. If your computer goes haywire and the IT person comes to help you – even though it’s her job and she’s obviously getting paid – I bet you’d feel some gratitude. Certainly there’s not always a place for it in these clear market exchanges, but I suspect that in many of them it still creeps in. I might even speculate that these are the kinds of exchanges where the instrumentality boost is particularly potent – that is, where the biggest drop off will occur post-task. These are the exchanges where the gratitude really surges when you need that person and then drops off once the exchange is done. When you’re getting help from someone at work, ever think about sending that person chocolates? Ever go through with it? If you said ‘yes’ and then ‘usually not,’ I suspect it’s in large part because once that goal is completed, then you’re on to the next thing. Wait, what was the first question again? Just kidding. But to avoid posting a whole separate paper in response to your two questions, I’ll break this up and get back to the potential communicative role of gratitude soon, in a subsequent post.

• Hugo Mercier 14 March 2011 (23:58)

  Thanks for your answer Ben, very illuminating. The example of the chocolate is right on target I think. It's still tempting to argue however that in most of the cases in which you feel gratitude for people doing their job, they have gone 'above and beyond' their duty. Or even if they haven't, one could argue that many situations in the workplace (like the IT guy coming to fix your computer) are not immediately, or intuitively market-like: it's not as if you were going to give him money on the spot. (Here's a potential experiment: compare the workplace IT guy with a guy you pay to come fix your computer. Maybe you'd feel more gratitude towards the first one?) Another explanation would be that the benefits for you can be huge (saving all the data on your non-backed up hard-drive), so that even if you pay the IT guy, there's an extra benefit that you feel requires gratitude (as opposed to, say, paying him to fix a trivial problem on your computer). Looking forward to the answer to the second (well, first) question!

• Ben Converse 16 March 2011 (19:33)

  You raised some interesting thoughts about the potential communicative role of the instrumentality boost. I think it’s fair to carve out two variations on the thought. I’ll argue against the first, but I think the second brings up some nice questions for future research. The first possibility you raise is the sort of mundane explanation based strictly on clarifying the stakes of the exchange. You wondered if both parties are clearer about what the costs and benefits are after rather than before, so there is less need to communicate this with gratitude after. I don’t necessarily accept the assumption that stakes or expectations are any clearer after than before. The best evidence we have against that probably comes from the follow-up we reported following Experiment 1. (Admittedly, there weren’t enough details there for anyone to anticipate the point I’m about to suggest.) In this study, captains and assistants worked side-by-side for the duration of the task, and the task was very simple. Here it would have been perfectly clear to both parties what was needed, what costs the assistant incurred, and what benefits it provided the captain. I don’t have a measure of this, but, if anything, it should have been more clear during the task (“active goal condition”) than after the task, when they had moved on to something else (“completed goal condition”). Even at a more conceptual level, I’m not convinced the role of gratitude is to really make clear what the helper needs – That seems outside of gratitude, in the “negotiation” or request and planning phases. That said, I agree with you that it is intriguing to think about a communicative role of gratitude. I agree that in our studies, it’s generally not a direct possibility. In Study 1 and some of our other, unreported studies, this would not be a possibility because the two parties simply do not have an opportunity for direct communication. But, I agree this doesn’t preclude the possibility that it’s part of the fundamental nature of gratitude. Indeed some recent papers are starting to unpack what communicative role gratitude actually plays. Adam Grant and Francesca Gino (2010, JPSP) show, for example, that expressed gratitude motivates prosocial behavior by increasing helpers’ feelings of social worth, that is, making them feel valued. Also, Nathaniel Lambert and Frank Fincham (2011, Emotion) have a brand new paper showing that expressed gratitude leads to more relationship maintenance behaviors. So there does seem to be a communicative benefit of experienced, and then expressed, gratitude. The next question is how that might fit into the effects we have observed. If it does contribute to the instrumentality boost, I think it would be through attempts to motivate or coax the helper. Expressing some gratitude during the task might be like dangling the proverbial carrot, as a way to urge them on. (This could increase the social worth, for instance, as suggested by Grant & Gino.) I would view this as entirely consistent with our proposed instrumentality mechanism. A beneficiary is more likely to coax a helper along when she needs the helper, that is, when the helper is instrumental. The mediation we observe in Expt. 2 is consistent with this. In any event, whether communication needs, per se, are contributing to this effect is an open question – we don’t have any direct evidence. I will note, however, that if, “in reality,” beneficiaries are routinely communicating their gratitude more before than after completion, helpers seem to be missing the signal. We show this in Experiment 2 in the current paper, where they expect no difference in gratitude. We also find in a couple of other studies (unreported here) that helpers often expect the exact opposite pattern of gratitude – They think they’ll be more appreciated after than before. This isn’t to say that beneficiaries couldn’t be botching their communication attempts, but it does suggest that the communication patterns may not be in the direction the communicative mechanism would require.

• Hugo Mercier 17 March 2011 (17:42)

  Thanks for this very interesting reply. Regarding the first point, I guess I was not trying to say that gratitude was meant to indicate what the receiver wants, but the fact that she's willing to incur a debt, but that is addressed in the rest of your reply. I'd have to think more about it, but you're right that the evidence you present seems to go against a communicative hypothesis. Again, great stuff, I really enjoyed the studies!