Guest Post by Amitai Shenhav, Princeton University
Tonight, after dinner, I will go out for ice cream at one of my favorite spots in Princeton. I will salivate in anticipation of my visit, delighting in all of the options that await me. I will carry that excitement with me as I enter the shop and examine all of the flavors I have at my disposal. And then, even as I’m thinking all of these good thoughts, I will quite predictably bring on board another feeling: stress over which flavors I should choose.
Decades of research have now shown we are often in a battle with ourselves when it comes to choice, even when it’s a choice between things we want – a “win-win” choice. We want choice and don’t like when people take it away from us, we want as many options as possible, and we feel good imagining all of those options. But we have a whole other set of responses to the process of making a choice. We vacillate, we experience anxiety, and we occasionally come up with ways to defer or avoid the choice altogether.
My dissertation advisor Randy Buckner and I performed a set of experiments, recently published in the Proceedings of the National Academy of Sciences, to try to understand what might be driving our paradoxical relationship with choices. How does our neural circuitry give rise to simultaneous experiences of anxiety and positivity when faced with only good options? And why are we still drawn toward such choices despite the difficult process that is likely to ensue?
To help address these questions, we asked people to make choices between pairs of products – such as school paraphernalia, gourmet snacks, kitchenware, and electronics – while we scanned their brains using fMRI. An auction-like procedure before the scan allowed us to individually tailor these choices to vary how much that participant would want each item individually, and how similarly they wanted the two items relative to one another. Our participants also knew that they would actually receive one of their choices, so they were motivated to choose the item from each pair that they wanted most. When they got out of the scanner, they told us how good and how anxious they had felt when facing each set of options.
These emotion ratings revealed that choices involving similarly high-valued options (such as an iPod shuffle versus a digital camera) simultaneously generated the most positive and the most anxious feelings. In fact, we found that any time people were choosing between products they wanted similarly, these two feelings were closely linked: the better they felt about their options, the more anxious they would feel about choosing between those options.
It may seem puzzling that a decision-making system that alerts us to high-quality options in our environment and enables us to compare those options to one another would simultaneously give rise to such disparate experiences. Our fMRI results help solve this puzzle. We found that these positive and anxious feelings were associated with separate brain circuits – each involving interconnected regions of the medial prefrontal cortex (mPFC) and a subcortical structure called the striatum – that have been implicated in different decision-making functions.
The circuit associated with positive feelings – comprising ventral (lower) regions of the mPFC and striatum – has traditionally been tied to the reward someone can expect from taking an action. The circuit associated with anxiety – comprising dorsal (upper) regions of striatum and mPFC – has traditionally been tied to the cost of taking that action, including both physical costs (e.g., how much will I have to strain my muscles?) and mental costs (e.g., how hard will I have to think about this?).
For win-win choices, there are two kinds of closely related costs that this dorsal circuit might be tracking. The first is the opportunity cost, or the value you give up by making a given choice. For example, if I choose the iPod over the camera, I’ve lost any of the value I could have gained by taking home the camera instead. However, in a separate study we found that anxiety continues to increase when faced with even more options of high value (e.g., iPod vs. camera vs. camcorder), even though in this case you still only ever give up one alternative by choosing the iPod. While this may just mean that anxiety reflects the sum total of all lost opportunities from making a choice, another possibility is that this feeling reflects a second kind of cost: the cognitive effort we must exert to settle the conflict between our different options (conflict that grows with more options).
Our results suggest that win-win choices feel both positive and anxiety-inducing because we have separate brain circuits simultaneously determining the value of potential rewards versus the costs or demands inherent to choosing amongst them.
This type of cost would explain why the dorsal mPFC, a region regularly associated with decision conflict, was among those that we found associated with anxiety. It may also explain why activity in this region around the time of the choice predicted whether the participant would later change their mind about that choice – even though the participants didn’t know they would have an opportunity to do so. This finding is consistent with the notion that this region is continually tracking signals that may demand additional cognitive resources after a choice is made.
Collectively, our results suggest that win-win choices feel both positive and anxiety-inducing because we have separate brain circuits simultaneously determining the value of potential rewards versus the costs or demands inherent to choosing amongst them. This helps make sense of why we experience both emotions rather than some sort of average of the two.
But still, one has to ask: If we are able to ascertain how anxious a choice will make us feel, why do we still invite such feelings upon ourselves by preferring more rather than fewer good options? Here we are left only to guess, but another of our findings may at least offer a clue.
As I mentioned earlier, we found a reward-associated network linked to positive feelings about win-win choices, separate from the network that was associated with anxiety for these choices. It turns out we were able to identify an additional set of reward-associated regions that responded to win-win choices , but responded to them differently than the positive affect-related network. Based on the different profiles of responses in these networks in our study and the previous literature, we speculated that this third network may be the one actually comparing the different options (determining which ice cream flavor is best given my current circumstances), while the network associated with positive feelings may respond more reflexively to the potential rewards (signaling that there are some delicious options ahead of me).
Our interpretation of this last set of findings is still speculative, and some of the relevant observations were made after our initial planned analyses so replication is needed before placing too much weight on them. Nonetheless, this is something we find intriguing and are currently following up on. If true, it would suggest that somewhat separate neural circuits may be in play when we are actually inside the store making a purchasing decision (when the process of comparison and attendant anxieties are most likely to come online) than when we are only leafing through catalogs or window shopping (and possibly deciding whether to enter the store in the first place).
If our initial approach-oriented decisions are partly guided by reflexive reactions to exciting possibilities that could result, it would make sense that such reactions would guide us to win-win choices that offer the most imaginary fodder – the most ice cream flavors or job options – before those options get kicked off to the systems that do the work of comparing them. After all, how bad can it be to choose between so many good options?
Taken together with previous research on this topic, our findings highlight the two sides of the emotional coin when it comes to choosing between things we want, while also helping to explain how we might have ended up with such a paradox. We hope that in better understanding the neural systems at the heart of this phenomenon we can eventually better understand how to manage the delicate balance between the upside and downside of win-win choices, in a world where our options for what to buy and how to spend our time are ever-increasing. For the time being, though, I have 10 different flavors to agonize over.
Are you a member of CNS with an interest in blogging? Consider contributing a guest post about your work or trends in the field. Email your ideas to CNS Public Information Officer, Lisa M.P. Munoz (email@example.com).