Social Pain: A Conversation with Naomi Eisenberger

Here is a new Conversations at Edge with Naomi Eisenberger, a professor in the Social Psychology program at the University of California, Los Angeles. She is director of the Social and Affective Neuroscience laboratory as well as co-director of the Social Cognitive Neuroscience Laboratory.

Dr. Eisenberger was featured in an earlier Conversations at Edge on Why Rejection Hurts [7.6.11].

Social Pain

A Conversation with Naomi Eisenberger [9.10.14]
Topic: Conversations

When I think of the work on social pain, and showing that some of the same neural regions that are involved in physical pain are involved in social pain, that can be very validating for people. For anyone who's felt the pain of losing somebody or who's felt the hurt feelings that come from being ostracized or bullied, there's something very validating in seeing this scientific work that shows it's not just in our head. It is in our head because it's in our brain. It's not just in our head, there is something biological going on that's interpreting the pain of social rejection as something that really is a painful experience. 


NAOMI EISENBERGER is a professor in the Social Psychology program at the University of California, Los Angeles. She is director of the Social and Affective Neuroscience laboratory as well as co-director of the Social Cognitive Neuroscience Laboratory.

Naomi Eisenberger's Edge Bio Page

SOCIAL PAIN

The kinds of questions that I'm interested in have to do with the feelings that come up in our closest social relationships. I've been on both sides of this equation, the negative side and the positive side. I've been very curious about both the painful feelings that we have when we lose our closest social relationships or we feel disconnected from others, as well as the pleasurable feelings that we have when we feel close and connected to those that we love. One of the things that we've been exploring in my Lab at UCLA is where those feelings come from. One way that we've been doing this is by turning to the brain to try to understand those feelings of social closeness or social distance.

We started off on the negative side. I started off with this question of: why do people talk about feeling hurt by social rejection? Why do they talk about the pain that comes up when they feel left out or excluded? It's a very palpable experience. Most people, if you ask them to think back to some very painful event in their life, instead of bringing up something that involves a broken bone they’ll often bring up an experience that involves losing a social relationship or being rejected. We all have these memories of being left out on the school playground or fears of being left out on the school playground. So I was very curious. Where do those fears and where do those feelings come from?

(29:23 minutes)

 

To explore this we essentially created a setting, created a situation that we could expose our subjects to, see what's going on emotionally when people feel rejected, and then what's going on in their brains when they experience this state. In this first study we brought people in to the fMRI scanner and had them play a virtual ball tossing game with two other supposed subjects. They thought they were playing this interactive game together. It turns out they weren't actually playing with two other people, but they didn’t know this yet. What happened during this game is that these two computer players stopped throwing the ball to our subjects so we could look at the brain and see what's going on for these individuals, what's going on in their brains when they're being rejected compared to when they're being included in the game.

What we found here was really interesting, that some of the same regions of the brain that process the painful experience—the distressing component of physical pain—are the same regions that seem to activate when people feel socially excluded. It gives some weight to this idea that rejection really can be painful. The same regions that process the unpleasantness of physical pain process the pain of social exclusion. This got us thinking about the overlap, the possible shared neural circuitry underlying physical and social pain. Maybe the same regions that process physical pain have been borrowed over the course of our evolutionary history to process social pain.

When you think about us as a social species, this actually makes quite a lot of sense. We rely on people, we need to be close to others, especially early on for protection, for care, for nourishment. To the extent that being separated from a caregiver or from a close other is such a threat to our survival, then actually feeling this pain signal when we're separated may be an adaptive way to prevent being socially separated.

We've gone on to explore some of this shared circuitry. We've tested a few ideas that come from it. One idea is that if physical pain and social pain really rely on some of the same neural regions, then one consequence of this is that people who tend to be more sensitive to physical pain should also tend to be more sensitive to social pain, and we've been able to show this in a few studies. We found that subjects who, at baseline, are naturally more sensitive to physical pain are the ones who later on report feeling more rejected by the same virtual ball tossing game where they get excluded.

We've also seen some genetic evidence for this. We find that people who carry the more rare version of the mu-opioid gene, which is linked to a greater sensitivity to physical pain (opioids are potent painkillers), are individuals who have a genetic disposition to be more sensitive to physical pain. These are the same individuals who report feeling more upset by social rejection; they show greater pain related neural activity in response to social exclusion.

A second consequence of this overlap that we've been exploring is whether certain factors that alter one kind of pain can alter the other in a similar manner, and probably one of the most interesting studies we've done is one where we looked at acetaminophen. We typically think of acetaminophen as a physical painkiller. In this particular study, we randomly assigned people to either take it everyday for two weeks or take a placebo everyday for two weeks. Instead of measuring their physical pain, we measured their social pain. We asked them each evening to rate their hurt feelings. We also then brought them in at the end of a separate study to look at their neural sensitivity to social exclusion. What we found is that the people who were taking acetaminophen reported less hurt feelings than people who were taking placebo, and they showed less pain related activity to social exclusion, just as a function of taking acetaminophen. We see this crossover effect in some ways, that this agent, which known to reduce physical pain, also seems to affect social pain.

Those are some of the things that we've been looking at on the more negative side of social experience. I've also become very interested recently in exploring the positive feelings that come from our social connections, and these have probably been the trickiest to really emulate in the scanning environment. It's easy to put somebody into a negative state when they're laying alone in a contained structure. It's harder to really get people to feel connected to their close others when they're all alone in the dark in the fMRI scanner.

In the same way that we've been curious about some of the neural substrates that might have been borrowed to support our experience of social pain, we've also been interested in what neural substrates might have been borrowed to support those pleasurable, warm feelings that we have when we're feeling connected.

One kind of substrate that we've been really interested in are those neural substrates that process temperature. The reason for that is because a lot of times when we talk about our feelings of closeness or connection, we talk about warmth. We talk about somebody making us feel warmhearted, we talk about our warm feelings toward somebody, so we wondered whether some of the same mechanisms that process warmth—that lead us to feel sort of pleasantly warm about a physical object—are the same mechanisms that lead us to feel warm about somebody else.

In one study we brought people into the scanner, and we wanted, again, to look at overlapping neural activity between physical warmth and social warmth. To look at physical warmth, we have them holding onto one of those warm packs that athletes will use where they crack them open and shake them up and it produces warmth in the packet. We scanned people when they were holding warm packs and a neutral temperature pack, and we also scanned them while they were experiencing social warmth. To do this we had the participants’ family members and friends, before the scanning session, write email messages to the participants. These were loving, tender messages that the subjects saw for the first time when they were in the fMRI scanner. We looked at what the brain was doing when subjects were viewing these socially warm images and whether or not it overlapped with what was going on when they were feeling these physically warm packs.

What we found, first just in terms of their self reports, not surprisingly, people felt more warm when they were holding the packs. They also reported feeling more warm when they were reading those nice messages. What was also interesting is that subjects not only reported feeling more connected when they were reading those messages, they also strangely enough reported feeling more connected when they were simply holding those warm packs.

The last thing that we saw was neural activity in reward-related regions during both tasks, and it turned out that there were several regions that showed overlapping neural activity. Some of the same regions that are processing physical warmth and the pleasantness of that sensory experience were the same ones processing the social warmth that people are getting from these loving messages.

Another line of research on the positive side is our research exploring the neural substrates of support-giving. Essentially there's been a lot of work showing that social relationships are critical for health. Most of the time, though, when people ask why social support is critical for health they assume that it's because of all the support that we get from others. Less often have people really looked at the support that people give to others. So when you help out a friend, or you take care of somebody, or you offer to do a favor for a family member, this is not considered support really because it's not helping us. We're doing things for other people. But this actually may be a key ingredient helping to explain some of the relationship between social support and better health.

We've been interested in some of the neural underpinnings of this particular state. So we ran a study where we brought in couples, and the female member of the couple was in the fMRI scanner, and essentially we scanned her brain while she was providing support to her partner. Her partner stood just outside of the fMRI scanner, and on certain trials he received electric shock. The female could support him on some trials by holding his arm as he went through this experience. This was a form of giving support to help somebody going through something negative or something painful.

There were two main findings here. The first is that we saw reward-related activity when people were providing support to somebody else. Some people say this is not terribly surprising. We all know that something about support-giving feels good. Some people might say that seems surprising because we're actually doing work to help somebody else, but we did see reward-related activity. We actually saw more reward-related activity when the females were touching their partners when they were getting pain—when they were support-giving—compared to when they were just touching their partners and their partners weren't getting pain. It seems like maybe there was something more rewarding about being able to provide support than just being able to be in physical contact with your partner when they're not going through anything negative.

The last interesting finding was that the females who showed more reward-related activity when they were support-giving were also the ones who showed less activity in the amygdala. This is a region that's involved in a lot of different things, but one of the things that it's involved in is processing threat. People who showed more reward-related activity that were involved in support-giving seemed to be showing less threat-related activity

If you look into animal work and if you think about the evolutionary aspects of this, it makes some sense. The idea here is that to the extent that we're in a caregiving situation, we need to remain calm. There may be some circuitry in place that turns down our own threat sensitivity so that we can engage in adaptive caregiving towards others. There may be something about caregiving that actually turns down our own internal stress level so that we can engage and provide adaptive help to others.

We've been trying to build on this and to look at some of the health consequences of support-giving. Is there something about support-giving that's actually stress reducing for the person who's giving the support? Not just for the person who's receiving it, which is what's typically been looked at, but for the person who's doing the helping. We have some preliminary data where people write a helpful note to a friend in need. After doing so, they go through a stress task. Those people that wrote the helpful note, as opposed to people who wrote just a control note, actually showed a calmer physiological profile to the stress task. They were showing less of a heart rate increase, less of a blood pressure increase, just as a function of writing this helpful message to a friend.

~ ~

I've always honestly felt a bit like a mutt in terms of the fields that have influenced me. I would call myself now a social neuroscientist, but I was trained as a social psychologist with probably a large influence from health psychology as well. Unlike most social psychologists, I tend to read a lot more animal work, and that's probably from health psychology, which borrows a lot from animal models of disease and how different kinds of social environments can affect health outcomes. I don't think it's a field, but people who have studied social relationships using animal models, as well as human models, have also greatly influenced me.

Individuals who have influenced me include Harry Harlow, John Bowlby, and Jaak Panksepp—people who have focused on what makes up the glue that binds people together. You have Harry Harlow looking at very basic attachment processes...what does the monkey prefer? Does he prefer the food or does he prefer the warmth, or the softness? You have people like John Bowlby who's trying to chart out a mechanistic way for understanding attachment, a child’s attachment to its mother. Then you have people like Panksepp, who use animal models to understand basic motivational systems like love, attachment, rage, caregiving. Probably the central theme is people focused on social relationships, and not just romantic relationships.

There's a whole wing of social psychology that focuses on romantic relationships, and in some ways that's why I never consider myself a relationship researcher. I'm interested in more than just the romance. I'm interested in the friendships, I'm interested in the parent/child relationships, all of that. I've been probably more influenced by the folks who’ve studied social connection from humans to animals.                                       

I think romance is interesting and a lot of people study romance. There's a lot of other meaningful types of relationships out there that people, for some reason, seem to be less interested in. So when I think about the warm feelings that I have for my employees or my close friends or my son, I want to know where those come from.                 

~ ~

How do I know what I'm doing makes a difference? I have two answers. Some of my research makes a difference because it reveals something to people that they probably already knew but maybe were afraid to believe. When I think of the work on social pain, and showing that some of the same neural regions that are involved in physical pain are involved in social pain, that can be very validating for people. For anyone who's felt the pain of losing somebody or who's felt the hurt feelings that come from being ostracized or bullied, there's something very validating in seeing this scientific work that shows it's not just in our head. It is in our head because it's in our brain. It's not just in our head. There is something biological going on that's interpreting the pain of social rejection as something that really is a painful experience.

There is something inherently interesting about figuring out how people work, figuring out where these warm feelings come from, and I'm not sure in the end if it will help anyone. That line of research is almost meant to just understand these things and maybe not meant necessarily to help people in the end.                 

I was the kind of person that always wanted to know what I wanted to do, but I never did, and I happened upon science in a way. I met a mentor who was gracious enough to take me on. We did a project together, and I found something new, and literally it was that excitement of discovering something for the first time that led me to want to be a scientist.

This still drives me to this day, this moment of discovering something new about human experience that we didn’t know before, the "Ah-ha!" moments. That's what reinforces things, and perhaps that's a selfish reason to pursue science, but for me that's certainly part of it, that feeling of discovering something new.

~ ~

I got my Bachelor’s of Science at UCLA. I majored in psychobiology. It was there that I got my first taste of research, so I wound up doing an Honors Thesis with Margaret Kemeny, who's a health psychologist and also a psychoneuroimmunologist. Because of our work together I think I realized I really wanted to do science. I really wanted to be able to ask questions about human nature, about human experience, and then use the data to answer those questions. After finishing my Bachelor’s, I went to UCLA and got a Ph.D. there. I worked with Matt Lieberman (now my husband), with Shelley Taylor and with Shelly Gable. And I think at the time when I was in my second or third year of graduate school I was introduced by Matt Lieberman to this new area of science called social neuroscience.            
                              
Before that I was probably more of a health psychologist. I was very interested in why social ties are so important for health, and there were all of these fabulous demonstrations of people who have more friends live longer and they are less likely to get sick. But I always felt like there was something missing with those models, and to me that was what's going on in the mind that translates the social world into whatever happens downstream in the body.              
            
I became very interested in social neuroscience as a way to connect that outer social world with the inner workings of the body to try to better understand some of the links between social issues and health. I took whatever classes I could, talked with whoever I could to learn all these techniques of social neuroscience. It was during that time that I, with Matt, did the study on social pain, looking at Cyberball, looking at people getting excluded in the scanner. My work built from there, continuing to look at that idea that physical and social pain overlap.           
                               
In some ways I veered off to get training in social neuroscience. During my post-doc I came back to health again. So during my post-doc I worked with Michael Irwin who's a psychoneuroimmunologist, and here I did something pretty different than what I did before. I ran a pretty involved study where we injected people with an agent that triggers an inflammatory response to look at how that inflammatory response affects people emotionally and affects people socially and also what's going on neurally as a function of that.   

~ ~

I have two main thoughts about differences between males and females when it comes to academics. When I was in graduate school I never thought about gender differences in how I was treated versus how anyone else was treated. It wasn’t until I became an assistant professor tha I started to notice things. It was subtle things, like I felt when a male colleague presented on my data they didn’t get as many questions. When I presented on my data there were a lot more questions. I don’t mean questions about clarity, questions to clarify. I mean doubtful questions, like, "I don’t believe what you're saying." Those kinds of questions. I've spoken with other female colleagues who have said the same thing. If she gives the exact same lecture as her husband to a class of students, she finds that she gets a lot more pushback than her husband.

There's also social networking things. Matt talks to other males, I talk to other females. The males that he talks to, because I guess they're males, tend to be in more powerful positions. It's oftentimes males who are department chairs. So it's Matt that's talking about jobs, about people moving, and orchestrating things. He's the one who's networking and pushing for positions in different places.

That's one difference. Another thing that I've noticed, I've seen more of this in my students, is that the way that science is done is more male friendly. This is my opinion, so some people may disagree with this, but I have a lot of female students, and when they go to conferences and give talks and people ask them questions that are challenging, that are maybe mean spirited, it makes them want to disengage. Sometimes, for male students, they see this as an opportunity to engage, to fight back...it's fun to argue. I’ll watch two males fight it out over some scientific question and think, "Oh my God, they're never going to speak to each other again," and meanwhile, they're having a really fun time. So there's something accepted about the way science happens that's a bit more male-oriented than female-oriented, in terms of its confrontational aspects.             

The last thought that comes to mind is that I've often wondered if females are less interested—at least some females—in self-promoting. There is maybe something uncomfortable about this kind of activity, or that they're just less interested, that it's not as important to them to get their message out, to be heard by a lot of people, and that's certainly something that I've dealt with myself. I like to do my science. I like to discover things. I like for other people to know about it, but do I want to go out and have speaking gigs all the time and get in front of audiences? Something about that makes me uncomfortable. I don't know if that's a "me" thing or if that's a female thing. "Me" and female are confounded so I can't really pull those two things apart.                                                                          

Stanislas Dehaene - Advances in Understanding the Signatures of Consciousness

The Edmond and Lily Safra Center for Brain Sciences hosts every Thursday Neuroscientists from around the globe to present their recent study. Earlier this year, Stanislas Dehaene gave a talk on the work of his lab in understanding the signatures of consciousness, the distinct markers of brain activity that correlate with subjective reports of conscious experience.

Stanislas Dehaene is a professor at the Collège de France, author, and (since 1989) director of INSERM Unit 562, Cognitive Neuroimaging. He has worked on a number of topics, including numerical cognition, the neural basis of reading and the neural correlates of consciousness. 

Advances in Understanding the Signatures of Consciousness

Published on May 1, 2014
Talk given on February 27, 2014

• A lecture given by: Stanislas Dehaene, Experimental Cognitive Psychology - Collège de France
• On the topic of: "Advances in understanding the signatures of consciousness"

Recap of Cognitive Neuroscience Society’s Annual Meeting (Scientific American Mind)

Here is a summary of some of the research presented at the 2014 Cognitive Neuroscience Society Annual Meeting. Daisy Yuhas at Scientific American Observations blog does the summarizing.

Brains in Boston: Weekend Recap of Cognitive Neuroscience Society’s Annual Meeting

By Daisy Yuhas | April 8, 2014



Greetings from Boston where the 21st annual meeting of the Cognitive Neuroscience Society is underway. Saturday and Sunday were packed with symposia, lectures and more than 400 posters. Here are just a few of the highlights.

The bilingual brain has been a hot topic at the meeting this year, particularly as researchers grapple with the benefits and challenges of language learning. In news that will make many college language majors happy, a group of researchers led by Harriet Wood Bowden of the University of Tennessee-Knoxville have demonstrated that years of language study alter a person’s brain processing to be more like a native speaker’s brain. They found that native English speaking students with about seven semesters of study in Spanish show very similar brain activation to native speakers when processing spoken Spanish grammar. The study used electroencephalography, or EEG, in which electrodes are placed along the scalp to pick up and measure the electrical activity of neurons in the brain below. By contrast, students who have more recently begun studying Spanish show markedly different processing of these elements of the language. The study focused on the recognition of noun-adjective agreement, particularly in gender and number.

Accents, however, can remain harder to master. Columbia University researchers worked with native Spanish speakers to study the difficulties encountered in hearing and reproducing English vowel sounds that are not used in Spanish. The research focused on the distinction between the extended o sound in “dock” and the soft u sound in “duck,” which is not part of spoken Spanish. The scientists used electroencephalograms to measure the brain responses to these vowel sounds in native-English and native-Spanish speakers. The Spanish speakers responded just like English speakers to the “dock” vowel sound, but not to the “duck,” sound, which was harder for the former group to identify. The finding is part of a larger body of research hinting at the possibility that vowel sounds like the o in dock that are produced on the periphery of the vocal tract are easier to perceive than the soft u sound produced in the middle of the vocal tract. By identifying these kinds of preferences, the researchers hope to better train language-learners to attune to sounds beyond their native tongue’s typical repertoire.

Birth control does not appear to effect cognition, according to Lena Ficco and colleagues of the University of Massachusetts Amherst. In one of the few neuroendocrinological studies presented, Ficco investigated whether contraception containing ethinyl estradiol changed the mental map-making or verbal abilities of women who had been taking this form of birth control for several years. Both verbal and navigation tasks are supported by estrogen. But because ethinyl estradiol suppresses estrogen, Ficco wondered whether there might be cognitive costs in taking contraception. Instead she observed no differences between women on these pills and a control group of non-pill-using women during the low-estrogen phase of their menstrual cycle. In future, Ficco hopes to assess whether length of pill use in an older population relates to any cognitive changes.

Researchers at Notre Dame University have some preliminary evidence that alcohol can set your body’s internal alarm clock. It can, in fact, make a relaxing Sunday morning seem like a manic Monday. The group wanted to investigate how alcohol, a physiological stressor, would alter the body’s cortisol awakening response, in which a flood of the stress hormone cortisol peaks as a person wakes up. Earlier work has demonstrated that this response is tied to psychological stress, prompting earlier rising on weekdays or during other anxious time periods. The Notre Dame group found that college students who consumed about four drinks on a weekend evening would awaken the next morning with significantly higher cortisol levels than non-drinkers. In fact, the researchers suggest that the stress-inducing effect is similar to that produced by cortisol levels on a weekday.

Finally, researchers have worked out a new nuance of the sound-induced flash illusion. In this illusion, an individual will either see a flash of light and hear two beeps or see two flashes and hear one beep. The curious thing is that people will report seeing two flashes if they hear two beeps, and just one flash if they heard one beep. Scientists at the University of Milan-Bicocca in Italy decided to replicate this illusion with a slight twist. Instead of actually flashing lights, they asked their subjects to keep their eyes closed and then used magnetic fields to change the electric currents surrounding neurons and stimulate the occipital lobe (a technique called transcranial magnetic stimulation). This stimulation triggered phosphenes, or the sensation of a flash of light—albeit without any actual light flashing. Because of how these phantom flashes are induced directly in the visual system, this approach enabled the researchers to time the stimuli and response to decipher whether this illusion depended on visual processing shortly after encountering a flash or beep. Indeed, they found that their subjects would report seeing two phosphenes after the researchers had induced just one flash with two auditory beeps, provided the stimuli were presented within relatively short succession. This suggests that the experiences in the brain’s early visual cortex can be modulated by sound. Lead author Silvia Convento explains that this kind of sensory overlap likely reflects brain organization that was beneficial to our ancestors. Even if it introduces some errors, the linkage would allow our brains gather and organize information from the environment more rapidly.

The lectures and symposia thus far have brought together a nice mix of history and hypotheses. Jon Kaas, who studies the organization of the mammalian brain at Vanderbilt University, for example, presented some of his ongoing work mapping the possible “sub-regions” of the motor cortex. The idea was inspired in part by research done in 2009 by neuroscientist Michael Graziano at Princeton University, who demonstrated that by stimulating a specific location on the motor cortex of an anaesthetized monkey, the animal would carry out a behavior such as bringing its hand to its mouth. Kaas has since investigated this and similar behaviors —including grasping, reaching, and climbing behaviors— in several primate species, including galagos, squirrel monkeys and owl monkeys. His findings have led him to the conclusion that the motor cortex may be divided into functional sub-regions with a specific active purpose, that direct movement across the body, and that the organization of these sub-regions could be consistent across primates.

In a totally different vein, a symposium on developmental cognitive neuroscience reflected on how lessons in this field could potentially guide policy on education and the justice system. For example, Margaret Sheridan of Harvard University reviewed the most recent published findings from the Romanian Orphanage Study, which reveal how extreme deprivation brutally inhibits mental development. This research complements ongoing study of the cognitive struggles of children growing up in poverty. On a more positive note, there are interventions that could help these children close the gap. Neuropsychologist Helen Neville of the University of Oregon discussed how her group has succeeded in training young children to improve their attention skills, often through simple and fun activities that ask children to concentrate and ignore distractions.

One of the most moving moments of the conference was in the opening keynote address by MIT neuroscientist Suzanne Corkin. Corkin spoke about the legacy of H.M., a patient who underwent a procedure to prevent epileptic seizures that left him unable to form new memories. Corkin, who spent decades studying H.M., discussed how he helped illuminate the distinctions between different memory forms and where they are located in the brain (to learn more about H.M., check out the May/June issue of Scientific American Mind). But Corkin also discussed the man behind the initials, describing his gentle and remarkably upbeat disposition, given that he was repeatedly confronting a confusing, context-free present. Her talk included a poignant and powerful audio recording of Corkin and H.M. chatting in 1992. In the excerpt, H.M. professes to “not mind” all of the tests and studies, saying simply, “I figure what’s wrong about me helps you help others.”


About the Author: Daisy Yuhas is an associate editor at Scientific American Mind. You can follow her on Twitter, @daisyyuhas

~ The views expressed are those of the author and are not necessarily those of Scientific American.

From Frontiers in Psychology: Perception Science, this is an interesting literature on the topic of self-generated mental activity and mind wandering (one and the same). Their analyses demonstrates a dramatic increase in the term “mind wandering” from 2006, with a corresponding and significant cross-over of psychological investigations of mind wandering into cognitive neuroscience (particularly in relation to research on the default mode and default mode network).

The era of the wandering mind? Twenty-first century research on self-generated mental activity

Felicity Callard [1], Jonathan Smallwood [2], Johannes Golchert [3], and Daniel S. Margulies [3]

1. Centre for Medical Humanities and Department of Geography, Durham University, Durham, UK
2. Department of Social Neuroscience, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany
3. Max Planck Research Group: Neuroanatomy and Connectivity, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

The first decade of the twenty-first century was characterized by renewed scientific interest in self-generated mental activity (activity largely generated by the individual, rather than in direct response to experimenters’ instructions or specific external sensory inputs). To understand this renewal of interest, we interrogated the peer-reviewed literature from 2003 to 2012 (i) to explore recent changes in use of terms for self-generated mental activity; (ii) to investigate changes in the topics on which mind wandering research, specifically, focuses; and (iii) to visualize co-citation communities amongst researchers working on self-generated mental activity. Our analyses demonstrated that there has been a dramatic increase in the term “mind wandering” from 2006, and a significant crossing-over of psychological investigations of mind wandering into cognitive neuroscience (particularly in relation to research on the default mode and default mode network). If our article concludes that this might, indeed, be the “era of the wandering mind,” it also calls for more explicit reflection to be given by researchers in this field to the terms they use, the topics and brain regions they focus on, and the research literatures that they implicitly foreground or ignore.

Introduction: A New Era of Mind Wandering Research?

One fundamental feature of the human mind is that mental activity does not cease when the mind is unoccupied by external demands. Instead, we often have thoughts and feelings that are unrelated to events in the here and now – a capacity that depends upon our mind’s ability to self-generate both cognitive and affective phenomena independently of environmental input (Smallwood, 2013). Throughout the article, we use the term “self-generated mental activity” to describe the fact that such experiences are largely generated by the individual, rather than occurring in direct response to experimenters’ instructions or to specific external sensory inputs. We use this term to describe an overarching category that encompasses a variety of phenomena – including mind wandering, daydreaming, fantasy, task-unrelated thought, and stimulus-independent thought (SIT). It should be noted that these phenomena do not exactly map on to one another (for example, self-generated mental activity would include deliberate problem-solving, and would also include daydreaming and mind-wandering, which may incorporate non-volitional processes). Despite this heterogeneity, the term “self-generated mental activity” captures a common phenomenon: that associated conscious experience is relatively more dependent on the individual’s concerns, preoccupations and hopes (i.e., self-generated), rather than immediate perceptual input (i.e., perceptually generated).

The number of studies in this special Research Topic (“Toward a psychological and neuroscientific account of the wandering mind”), coupled with the variety of topics they address, make clear that there is currently significant scientific interest in self-generated mental activity. This was not necessarily predictable: even a decade ago, investigations of these experiences were relegated to the backwaters of psychological research (see Smallwood and Schooler, 2006 for a discussion). Indeed, psychologically oriented research on self-generated mental activity was hampered for much of the twentieth century, because of the powerful influence that behaviorism exerted for many decades (e.g., Watson, 1913 and Skinner, 1953; for reflections, see Callard et al., 2012). From the 1940s to the 1970s, researchers in the field perceived that topics involving self-generated mental activity were not greeted at all positively by many senior psychologists: the field was still dominated by the pervasive meta-theory bequeathed by Watson and Skinner, which resulted in the exclusive legitimacy of behaviorist methodologies in many departments and many peer-reviewed journals (Klinger, personal communication, 2013). Nonetheless, pioneering and still influential psychological research was conducted by a small number of researchers during these decades – in particular the path-breaking research on daydreaming by Jerome Singer and his doctoral students John Antrobus and Kenneth Pope (e.g., Singer and Antrobus, 1965; Antrobus, 1968; Pope and Singer, 1978; Singer and Pope, 1978), and subsequent research by Klinger (1971), as well as by, e.g., Giambra (1974, 1993). This research was frequently not published in the most prestigious psychology journals, and often appeared in monographs (e.g., Singer, 1966, 1975) or in smaller or speciality journals (e.g., Perceptual and Motor Skills (e.g., Singer and Antrobus, 1963) and Imagination, Cognition and Personality).

Those early works undoubtedly provided the foundations upon which isolated researchers continued to work in the late twentieth century (e.g., Einstein and McDaniel, 1997; Wegner, 1997). In this article we focus on the first decade of the twenty-first century – that moment during which research on self-generated mental activity moved out of the shadows, towards the scientific mainstream, and increasingly into journals with greater apparent scientific credibility. In contrast to earlier, widespread dismissal of or lack of interest in self-generated mental activity, many researchers now acknowledge that these phenomena have broad implications for many elements of psychological and neural function. For example, research has focused on how self-generated thought might be related to both physical (Epel et al., 2013) and mental health (Killingsworth and Gilbert, 2010); explored its implications for attentional control (Smallwood and Schooler, 2006; McVay and Kane, 2009; McVay et al., 2009; Smallwood, 2010); considered its implications for educational success (Smallwood et al., 2007a); and addressed its relation to psychiatric conditions such as depression (Smallwood et al., 2005, 2007c, 2009). A further strand of research has begun to illuminate how mind wandering is related to the nature and functions of intrinsic changes in brain activity (Mason et al., 2007; Smallwood et al., 2008; Christoff et al., 2009; Kam et al., 2011; Smallwood et al., 2013b). High-profile and/or highly cited publications in Science (Mason et al., 2007; Killingsworth and Gilbert, 2010) Proceedings of the National Academy of Sciences (PNAS; Christoff et al., 2009; Szpunar et al., 2013) and the British Medical Journal (Galera et al., 2012) make it likely that this trend will continue.

Prior to our research for this article, our sense was that “mind wandering” is now the dominant term used by researchers to characterize the self-generated mental activity in which they are interested. However, the history of psychology and proximate disciplines indicates a broader palette of terms relating to self-generated mental activity (e.g., fantasy, daydreaming; Callard et al., 2012). We were intrigued, therefore, by: (i) whether the term “mind wandering” currently does dominate research on self-generated mental activity; (ii) if it does dominate, when and how it came to do so; (iii) whether research on mind wandering, specifically, is closely related to research on other, analogous phenomena and experiences; and (iv) whether mind wandering research tends to focus on particular psychological phenomena and processes, and on particular brain functions.

Our previous research has highlighted the benefits of explicit reflection on emergent scientific fields. We have emphasized the value of analysing how certain assumptions can become (prematurely) embedded; how certain terms and concepts can become solidified over others; how normative claims (explicit or implicit) can be made about the phenomena under investigation; and how distinct bodies of research with different terminologies and ontological foundations can be brought together (or kept apart; Callard and Margulies, 2011; Callard et al., 2012; Margulies et al., 2013). This article combines our scientific and historical interests in self-generated mental activity with an exploration of how recent peer-reviewed research has described such experiences. We combine bibliometric analyses with our own expertise in the history and current state of research on self-generated mental activity. Our hope is to encourage reflection on these issues by other scientists who are building understanding of how the mind self-generates conscious experience. One central aim is to bring analytical visibility to the fact that Frontiers in 2013 has a special topic on the wandering mind, rather than, for example, on SITs, task-unrelated thoughts (TUTs) or self-generated cognition.

Methods and Results

We wished to investigate three different aspects of recent peer-reviewed literature that address self-generated mental activity: (i) temporal changes in a subset of terminologies used to describe self-generated mental activity; (ii) the particular research topics (e.g., particular psychological processes and/or brain functions) that characterize research currently being conducted on what we expected to be the most common category of self-generated experience: mind wandering; and (iii) where there is cross-fertilization of research interests and findings within the wide field of research on self-generated mental activity, and where there is compartmentalization of research that remains separated from other research arenas.

We determined that a combination of methods would allow us to address these three areas of inquiry: (i) we used quantitative and qualitative methods descriptively to explore recent historical changes in the use of terms for self-generated mental activity, and in the topics on which researchers investigating self-generated mental activity focus; and (ii) we performed a bibliometric analysis and visualized a citation and terminology map of the literature.

Descriptive Analysis of Terminology for Self-Generated Mental Activity

Historical Changes in Terminology to Describe Self-Generated Mental Activity
We searched the ISI Web of Science database in March 2013 in order to plot changes in terminologies used to encompass self-generated mental activity. We used our own expertise regarding historical and current research on mind wandering and related phenomena to identify the terms under which to search (these comprised: daydreaming, mind wandering, SIT, TUT, spontaneous cognition, spontaneous thought, and fantasy proneness). We then assembled a database of all articles that cited the three most highly cited articles for each of the terms. We did this because using the terms themselves would have restricted the pool of articles by a terminology, rather than investigating the pool of literature to which articles using those terms were contributing. We restricted our search to the decade prior to this special research topic in Frontiers (i.e., 2003–2012). This is a relatively short time span; however, it covers precisely that point at which we were aware that there appeared to be substantially growing interest on the part of psychologically oriented research communities in self-generated mental activity. The number of publications yielded by this procedure was then plotted over time (in relation to whether each of the original terms appeared in the title and/or abstract and/or keyword of each item in the database; Figure 1).

FIGURE 1  

FIGURE 1. Changes in the frequency of citations across the ten years prior to 2013. It is apparent that the term mind wandering has seen a rapid increase in the frequency of papers using this term over this period. By contrast, some research that uses other, related terms has remained at a relatively consistent level over the same period. The x-axis describes the year the paper was published. The y-axis describes the number of papers published in each year that have the term in the title, abstract, or keywords. Different colored lines describe the different terms used to describe self-generated mental activity.

What is particularly noticeable is the difference between those terms that hover at roughly the same level throughout the decade (e.g., fantasy proneness and spontaneous cognition) and those that become increasingly prominent (daydreaming, SIT and – particularly noticeably – mind wandering). Indeed, in 2010, there is a step-change in the prominence of research that uses the term mind wandering. (Mind wandering, noticeably, does not feature until 2006, subsequent to which it very closely tracks daydreaming until the divergence in 2010 precipitated by the much larger increase in mind wandering than in daydreaming.)

Changes in Research Focus within Literatures Addressing Mind Wandering

Data from the Section “Historical changes in terminology to describe self-generated mental activity” revealed the recent and growing dominance of research using the term mind wandering. In order to interrogate this literature more closely, we investigated changes in the use of keywords in the mind wandering articles (i.e., those articles in the database using “mind wandering” in the title, abstract or keywords) to understand any changes in research focus that have accompanied the rapid growth of interest in mind wandering.

As the number of papers was relatively small (n = 145), and the number of keywords was relatively large (approximately 900), we reduced these terms into superordinate categories (which were based on the substantial experience of one of the authors, Jonathan Smallwood, in mind wandering research). This was performed simply to reduce the number of categories to a manageable number. Based on an analysis of the distribution of these keywords, Jonathan Smallwood identified a broad set of categories of research (n = 15) that accounted for a large percentage of the terms used. These categories were selected on the basis of the distribution of keywords identified and largely served to reduce pseudonyms (e.g., the terms “default mode” and “default mode network” were collapsed into a single category) and to create meaningful psychological categories (e.g., recollection and working memory were collapsed into a category of memory processes). Some of these categories were labeled using categories introduced during the process (e.g., the keywords “resting state,” “gray matter,” “prefrontal cortex,” etc. were all grouped within the category “cognitive neuroscience”). A second rater (Johannes Golchert) independently assessed the same data using the set of categories produced by Jonathan Smallwood. We interrogated these data using descriptive statistics. Although these ratings were reliable, we make no claims that they reflect a definitive set of research categories; they simply serve to provide a smaller number of categories with which we can explore broad changes in recent research on mind wandering.

We were then interested in exploring two themes. First, in order to identify the large categories of research that are associated with mind wandering over the last decade, we plotted the relative proportion of each of the categories in our sample in the form of a pie chart (Figure 2). The largest category of keyword associated with papers that used the term mind wandering was the term cognitive neuroscience, which occurred over 25% of the time (and in which the subsection “default mode network” represented a significant proportion). The next largest sets of categories were: memory processes, attention and perception and performance. Notably, therefore, approximately 25% of the key words found in our sample were related to aspects of behavior that mind wandering has been shown to derail.

FIGURE 2  

FIGURE 2. Pie chart illustrating the different categories that form the focus of mind wandering research papers over the last decade. The categories were identified by one author (Jonathan Smallwood), and were derived from papers’ keywords. Their applicability was confirmed by an independent assessment of these categories by a second author (Johannes Golchert). Agreement between authors was high.

Our second aim was to consider historical trends that occurred in the use of key words over the last decade. We plotted the number of papers on mind wandering falling within each of our identified categories each year over the period of interest (Figure 3A). Given that the category of cognitive neuroscience accounted for over a quarter of our data, we plotted the historical trend in this category, and in one of its largest subcomponents, the default mode network, separately from all other keywords (Figure 3B). It can be seen that certain keywords show a pattern of slow and steady growth and are present in the majority of the years covered by our study (for example, SIT/TUT). Others have shown a rapid increase in their prevalence; of these, some were not present in the initial period (such as control), and others (such as consciousness) emerged concurrently with the turn to “mind wandering” in 2006. What is particularly noticeable is the predominance of cognitive neuroscientific research in the last 2 years of the selected period (2010–2012); within this same short time span, the visibility of research specifically on the default mode/default mode network is also striking.

FIGURE 3  

FIGURE 3. Changes in the categories that form the focus of mind wandering research papers between 2003 and 2012. It can be seen that almost half of the citations refer to psychological phenomena (A) whereas approximately the same amount refer to research focusing on cognitive neuroscience and in particular the default mode network (B).

Visualization of Research Literatures and Co-Citation Networks

The methods used above allowed us to gain a preliminary understanding of the rise of particular terminologies over the last decade, as well as the topics of enquiry being focused on within the mind wandering literature. We were also keen to have a greater understanding of the use of different terms employed in the field, the specific shifts that have occurred in the form of novel domains of investigation, and which communities cite – or do not cite – each other’s research. CiteSpace is a software tool that has been developed to map various aspects of citation networks, including the evolution of a literature over time (Chen, 2004, 2006). Based on databases of the scientific literature, it provides a flexible and interactive interface for assessing numerous aspects of the dynamics within a given field. For example, of relevance to the current analysis, CiteSpace enables the user to slice a database of literature into years of publication, and then to assess similarity of articles based on the similarity of referenced citations. Terms can also be culled from titles, keywords, and abstracts to depict their proximity based on shared inclusion in articles. The visualization platform then allows links between references or terms to depict the first year in which a connection occurred using colored edges. Subsequent usage of a term or reference can then be visualized using concentric colors that represent the frequency of citation (or use) for each year. While CiteSpace provides numerous further analytic possibilities, we constrained our analyses here (depicted in Figure 4) to the two analyses described above to facilitate interpretability.

FIGURE 4

FIGURE 4. CiteSpace was used to visualize the literature from 2003 to 2012 presented in Figure 1. The colors represent years, the squares are highly used terms (top), and the circular nodes are cited articles (bottom). Each colored circle represents the number of citations/uses during that year. Edge links between nodes represent co-occurrence (in the case of terms) and co-citation (in the case of articles), with the color representing the first year in which the connection was found.

The network visualization presented in Figure 4 provides information about the organization of the citation network and use of terms over the past decade (2003–2012) using the database procedure described for the Section “Historical changes in terminology to describe self-generated mental activity” (and presented in Figure 1). CiteSpace deals with the problem of substantial inconsistencies in the number of articles per year by including only the most relevant, which are determined by their number of citations. The top 1% of cited articles (with a maximum of 100) from each year were included in the similarity calculations, which were based on co-citations and common term use. The proximity of nodes in the graph represents this similarity, and the visualization also reveals specific terms with high frequency of use (Figure 4, top) and articles with high accumulation of citations (Figure 4, bottom; represented by node size, where color subdivides frequency by year). The colored edges represent the earliest year in which the connection was found.

The network visualization renders immediately clear the emergence of a new tightly clustered research field beginning in 2006–2007 (green colored edges) that is characterized by terms such as “default mode network” and “functional connectivity” (Figure 4, top). “Stimulus-independent thought” falls at the edge of this cluster, closer to the psychological literature from which it emerged, though still in close proximity to the cognitive neuroscience cluster. “Working memory” is situated as a bridge between this more recent field shift toward cognitive neuroscience, and a distinct cluster described by “fantasy proneness.” On the other side of this cluster (i.e., at a distance from the cognitive neuroscience cluster) lie a circle of related terms that tie together dissociation, hypnotic susceptibility, imaginative involvement and mystical experiences. Most recently, the term “sustained attention” emerges in a tight orange cluster from 2011 from the cognitive neuroimaging literature, suggesting the emergence of a novel field of interest in self-generated mental activity.

The cited references tell another aspect of the story, indicating the role certain articles may have played in providing links between various fields. For example, Smallwood and Schooler (2006), which was published in a major psychology journal, lies right at the heart of the cognitive neuroscience cluster, depicting the central role of that article in the emerging link between “default mode” and mind wandering that has characterized the second half of the last decade. (Notably, most of the highly cited cognitive neuroscience publications (Figure 4, bottom) address the default mode and/or default mode network, e.g., Shulman et al., 1997; Gusnard et al., 2001; Raichle et al., 2001; Greicius et al., 2003; Fox et al., 2005; Buckner et al., 2008). Meanwhile in the cluster that includes fantasy proneness, dissociation and hypnotic susceptibility, the Tellegen Absorption Scale (Tellegen and Atkinson, 1974) has overwhelming significance, which extends across the entire period under investigation.

Discussion

Our analyses revealed several important features of recent research on self-generated mental activity. First, there are a number of distinct terminologies and topics of research, and these are subject to differential historical changes in terms of the frequency with which they are used. Certain terms and research topics (those associated with SIT, task-unrelated thought, and mind wandering) have grown in stature over the last decade. Our initial conjecture vis-à-vis the growing prominence of the specific term mind wandering is upheld: the last decade has not only seen an increase in research on mind wandering, but has also been marked by a solidification of the use of this term over and above alternatives.

The specific term “mind wandering” becomes prominent only very recently; prior to that, it was terms such as daydreaming and task-unrelated thought that were more dominant. Indeed, it is conceivable that a paper published in 2002 by Schooler in the high-profile Trends in Cognitive Sciences, which focused on dissociations between experience and meta-consciousness, and which used the phrase “catching one’s mind wandering” in the abstract, helped to facilitate a shift towards the scientific community’s use of the term “mind wandering.” (Prior to 2002, there are few uses of “mind wandering” in the psychological literature; though see, amongst others, Giambra, 1989, which operationalized “daydreaming/mind wandering” through task-unrelated thoughts, and Einstein and McDaniel, 1997, which appears to be the first to use “mind wandering” in the title). In the decades prior to those we have focused on here, the most prominent psychological research on self-generated mental activity – carried out largely by Singer, Antrobus, Klinger and Giambra – privileged constructs and phenomena that included daydreams/daydreaming, fantasy and TUTs. It is possible to advance hypotheses about why the term mind wandering superseded some of these terms. For example, the construct “fantasy” could well have been regarded (by both contemporary cognitive psychologists and cognitive neuroscientists) as too closely associated with psychoanalytically inspired research, which was increasingly jettisoned from mainstream psychology as the twentieth-century proceeded. Nonetheless, more research needs to be done to more thoroughly understand why certain terms have been overtaken by others.

The rise in “mind wandering” research has been aided by its translation from cognitive psychology into cognitive neuroscience in the last half decade. Here, it has a close tie to research on the default mode network and functional connectivity. These are, notably, research fields that, like the term “mind wandering,” came to visibility in the twenty-first century (Callard and Margulies, 2011; indeed, it could be said that the mind wandering field and the resting state/default mode network fields appear to act as motors for one another – each raising new questions for the other field to answer, and each drawing interested researchers into one another’s orbit). The close link between mind wandering research and research on the default mode network arguably implies a one-to-one mapping between a kind of experience and a particular brain network, which is likely not doing justice to the varieties of self-generated mental activity, nor to the complexity of the neural processes that contribute to these heterogeneous states. Indeed, we have argued that this tight association between mind wandering and the default mode network is at least in part owing to its historical context. That cognition has been understood largely in relation to action and environmental influences meant that “mind wandering” (as the apparent opposite of such cognition) became bound to the activation of the so-called task-negative network (the default mode network). We have argued that “this apparent ‘see-sawing’ of neural activity between two widespread brain networks suggest(s) rather intuitive – and folk-psychological – distinctions between opposing psychological functions of goal-oriented cognition and spontaneous thought” (Callard et al., 2012).

We also found that certain key articles have acted to bring together psychological and neuroscientific perspectives on self-generated mental activity. For example, in 2006 Smallwood and Schooler published “The restless mind”; while this article did not use mind wandering in its title, it did have mind wandering as a keyword, and also contained the sentence: “By referring to this phenomenon as mind wandering, a term familiar to the lay person, we hope to elevate the status of this research into mainstream psychological thinking.” A year later, Mason et al. (2007) published their high-profile Science article: its title brought together mind wandering, the default mode and SIT. This continued in 2008–2010 through the publication of articles by Buckner et al. (2008), Christoff et al. (2009), and Andrews-Hanna et al. (2010), which further cemented the links between the default network and mind wandering. In 2009, Schooler and Kane, each of whom has published well-known and highly cited research on mind wandering, organized a large symposium on “Wandering Minds and Brains” at the annual meeting of the Psychonomics Society in Boston (http://www.psychonomic.org/pdfs/PS_Call_For_Symposia_B.pdf). This symposium is likely to have acted as a stimulus for additional research (and subsequently citations) on mind wandering amongst cognitive psychologists and those in related fields. Mind wandering has also played center stage to at least two commentaries in high profile journals relating to executive control (McVay and Kane, 2010) and its neural basis (Gilbert et al., 2007). In the same vein, Killingsworth and Gilbert (2010) used the term “a wandering mind” in the title of their high-profile 2010 Science publication. Together, this combination of high impact papers and theoretical controversies set the stage for the very rapid growth of research on mind wandering from 2010 onward.

Second, both from our analysis here and our knowledge of the longer history of research that investigates self-generated mental activity, it can be argued that there are a number of relatively discrete research communities that have investigated such activity – often under the umbrella of distinct terms. Figure 4 indicates that this burgeoning interest in self-generated mental activity within cognitive neuroscience has been relatively distinct since the second half of the last decade from significant areas of psychological research that have addressed related processes and phenomena. The lack of relationship between these research communities needs to be understood in part via a longer history in which the concept of “fantasy proneness” was formulated in the 1980s (originally by Wilson and Barber, 1981, 1983). Fantasy proneness is often treated as a trait (cf. mind-wandering, which has also been treated as a trait, e.g., Kane et al., 2007), was shown in certain respects to be correlated with certain psychopathologies, and was in many cases separated out from research on fantasy as process (Klinger et al., 2009). Indeed, researchers working on absorption, fantasy proneness, and dissociation (where much of the focus is on the maladaptive and/or psychopathological) tend to be relatively secluded from cognitive psychologists and cognitive neuroscientists researching mind wandering and daydreaming (where much of the focus is on processes that are commonly regarded as “normal”). This raises many interesting questions about the extent to which each distinct research sub-community might be using different kinds of normative assumptions and conceptual frameworks to understand related phenomena.

Implications

We suggest that the range of terms used to investigate self-generated mental activity raises an important question for future research. The literature is heterogeneous and complex, and more research is needed to understand the conceptual, methodological and phenomenological overlaps between the objects of study being investigated by these different research communities. Certain research topics have gained traction under the umbrella of mind wandering, while others might well take shape in a field focused on the investigation of fantasy or of spontaneous cognition.

Rather than regarding such trends as a passive result of collective research agendas, we contend that it would be valuable to explore the motivations and forces that provide traction for certain terms, constructs, and approaches at particular moments in time. What is to be gained by a field through turning its research toward a previously ignored phenomenon and/or construct? Do certain formulations or terms have more flexibility than others for engendering particular interdisciplinary overlaps and crossings that have recently taken place? What is gained and what is lost when researchers investigating maladaptive and/or psychopathological manifestations of phenomena are separated from other researchers focusing on other manifestations of the same (or related) phenomena, which are frequently assumed to be “normal”? What causes certain terms and formulations to be “overtaken” by others at particular historical moments and by particular scientific communities? And what are the consequences of certain research fields remaining immune to and isolated from other research fields? Would greater cross-fertilization bring new insights into each respective research community? These are all important questions to which the research community needs to devote more attention if it hopes to provide a comprehensive account of self-generated mental activity.

One general question that this line of research raises is which terms we as a discipline should use to describe the phenomena of self-generated mental activity. Terms such as “mind wandering” and “daydreaming” have attracted the attention of writers in high-profile non-peer-reviewed publications (e.g., Jarrett, 2009). Jonah Lehrer published an essay with the normatively explicit title “The virtues of daydreaming” in The New Yorker in 2012 (Lehrer, 2012); John Tierney published “Discovering the Virtues of a Wandering Mind” (Tierney, 2010a) and “When the Mind Wanders, Happiness Also Strays” (Tierney, 2010b) in 2010. These articles disseminated research that has in the last few years become some of the most highly cited in the field (including Killingsworth and Gilbert, 2010, as well as research by Schooler, Smallwood, and Christoff). Such non-peer-reviewed publications, by drawing on long-standing general cultural interest in daydreams and wandering minds, have undoubtedly contributed to building excitement and interest in and outside of the scientific fields. The concept of mind wandering, while highly amenable to public interest, is an umbrella term for many different aspects of cognitive experience, and is relatively poorly specified (cf. research that focuses on specific aspects of self-generated mental activity, e.g., certain properties of the state, such as stimulus dependence). Different theorists are interested in developing accounts of different aspects of self-generated mental activity, and disagreements can arise because theoretical accounts to describe different elements of these experiences are often seen as contradictory, when in fact they need not be (Smallwood, 2013). One important aim for a more comprehensive account of self-generated mental activity is to develop component process accounts of the different elements of the experience.

Our analysis also revealed how much mind wandering research has focused on what self-generated mental activity interrupts. Much mind wandering research is “negatively” driven, because of a focus on the costs of the experience, rather than on exploring the phenomenology of the underlying processes that drive the mind to self-generate experiences. One example is the role that mind wandering plays as a contributory factor to poor concurrent task performance (see for reviews, Smallwood and Schooler, 2006; Smallwood et al., 2007b; Smallwood, 2013). Recently, mind wandering has been drawn into new arenas of research, such as meditation (Mrazek et al., 2012, 2013). This research topic has perhaps developed because of research demonstrating that mind wandering has robust links to unhappiness (Smallwood et al., 2009; Killingsworth and Gilbert, 2010). Although this research is important, there are several aspects of this focus on the costs of mind wandering that are worthy of comment. Although self-generated mental activity can contribute to unhappiness and error, it can also be associated with creativity (Baird et al., 2012), future planning (Baird et al., 2011), and a tendency to make patient, long-term choices (Smallwood et al., 2013a). These are all important cognitive capacities, indicating that self-generated mental activity is associated not only with psychological costs. Given that self-generated mental activity is so common in daily life, and is coupled with both costs and benefits, it seems that a more nuanced view of the experience is warranted (for a discussion see Smallwood and Andrews-Hanna, 2013). Although understanding the costs that mind wandering can have in particular contexts is important, it is worth reflecting on whether this strong focus might have occluded other approaches, which do not start from a position of focusing on what the phenomenon of interest (mind wandering) interrupts or limits. One of the most self-evident facts that phenomena such as mind wandering indicate is how little we understand about how and why human minds engage in self-generated mental activity to the extent that they do.

In conclusion, our analysis highlights important disciplinary and methodological trends that have accompanied research on self-generated mental activity in the early twenty-first century. We hope to have made explicit the complex role that heterogeneous scientific communities (in their relations or non-relations with one another) can have in consolidating particular terms, methods and areas of enquiry in research on self-generated mental activity. Ultimately, such analyses may open up new approaches, as well as new connections between different research communities. If this is, indeed, the “era of the wandering mind,” it is appropriate that explicit reflection be given by mind wandering researchers to the terms they use, the topics and brain regions they focus on, the research literatures they implicitly foreground or ignore, and the research topics in which they do or do not embed their research. Such reflection will, we hope, help to resolve contradictions and impasses that currently hamper research, and accelerate the pace of research on the intriguing puzzle that self-generated mental activity poses to our research communities.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

Felicity Callard is supported by two Wellcome Trust Strategic Awards to Durham University (WT086049 and WT098455MA). The authors are grateful to the Neuro Bureau and to members of the Department of Social Neuroscience in the Max Planck Institute for Human Cognitive and Brain Sciences for ongoing creative input. The authors are grateful to the two peer reviewers, whose commentaries greatly strengthened the paper – particularly as regards the longer history of research on daydreaming and related phenomena.

Author Contributions

Daniel S. Margulies conceived of the paper; Felicity Callard and Jonathan Smallwood drafted the manuscript with contributions from Johannes Golchert and Daniel S. Margulies; Johannes Golchert and Jonathan Smallwood undertook the bibliographic analyses and ratings of keywords; Daniel S. Margulies conducted the CiteSpace visualization in consultation with Felicity Callard and Jonathan Smallwood. All authors read and approved the final manuscript.

References are available at the Frontiers site.