Abstract
The fields of psychology and neuroscience are in the midst of an explosion of research aimed at illuminating the human imagination—the ability to form thoughts and mental images that stretch beyond what is currently available to the senses. Imaginative thought is proving to be remarkably diverse, capturing the capacity to recall past experiences, consider what lies ahead, and understand other people’s minds, in addition to other forms of creative and spontaneous thinking. In the first part of this article, we introduce an integrative framework that attempts to explain how components of a core brain network facilitate interacting features of imagination that we refer to as the mind’s eye and mind’s mind. We then highlight three emerging research directions that could inform our understanding of how imagination arises and unfolds in everyday life.
Consider for a moment what life would be like without imagination. Our days would certainly be less inspired without the creative works of Dalí, Beethoven, or Austen. But these luxuries are just the tip of the iceberg. If our problem-solving capacities were diminished, breakthrough inventions would be nonexistent. Without the capacity to represent other people’s minds, our relationships as we know them would unravel. Without a mental timeline or the capacity to self-reflect, our sense of self may well dissolve. Indeed, we would be glued to sensory input from the present moment, yet we would arguably lack the capacity to understand the here and now and the motivation to savor or escape it.
In this article, we adopt a broad definition of imagination as “the power or capacity to form internal images or ideas of objects and situations not actually present to the senses” (Oxford English Dictionary, 2021)—a framework synonymous with terms such as perceptually decoupled thought, self-generated thought, or internal mentation. From this perspective, we first synthesize important findings from the recent surge of research in the psychology and neuroscience of imagination to propose an integrative neurocognitive framework, and we then discuss where the field is going (or perhaps ought to go) in years to come.
Mapping the Cognitive and Neural Landscape of the Imaginative Mind
The power of imagination lies in its ability to transcend the confines of the present moment to support a vast universe of mental content, spanning the real to the fictional, the self-focused to the social, the pleasant to the unpleasant, the past to the prospective, the specific to the general, and so on. In light of this complexity, a natural question is whether there are organizational principles that govern imaginative thoughts and whether such an organizational framework is reflected in neural tapestries of interacting brain regions (Abraham, 2020; Raffaelli et al., 2020). Many researchers have approached this question through different lenses, focusing on particular features of imagination such as its constructive, past, prospective, social, self, creative, or imagery-based properties. For instance, in the past decade, a great deal of research has been inspired by the idea that our ability to imagine specific events unfolding in the future (called episodic future thinking) may involve retrieving and recombining details from specific past events (called episodic memory; Addis, 2020; Schacter et al., 2017) around a scaffold of factual knowledge (called semantic memory; Binder et al., 2009; Irish, 2020). There also is greater appreciation that many forms of imaginative thought involve mentally constructing a scene (Mullally & Maguire, 2014), intersecting with a literature on mental imagery that has begun to expand from a focus on sensory regions to a focus on activity patterns across the entire brain (Pearson, 2019). The idea that we can “self-project” into the past or future or consider another person’s perspective has also highlighted the reflective and social qualities of imagination (Buckner & Carroll, 2007; Lieberman, 2007).
Here, we bring together many of these foundational views and suggest that there may be a unifying organizational framework for imagination. In this integrated framework, we position all forms of internally guided cognition under the umbrella term imagination, which we propose serves as the common function of a large-scale brain system called the default network 1 (Buckner et al., 2008). Building on similarities in subdomains of imagination, we suggest that there are two imagination-driven subsystems within the default network: one that supports our ability to evoke an internal mind’s eye and the other an internal mind’s mind (Fig. 1; Raffaelli et al., 2020; for related views, see D’Argembeau, 2020; Irish, 2020; Sheldon et al., 2019).
Imagination in the eye and mind of the beholder. We propose that imaginative thought is an outcome of the individual and collaborative inputs of the mind’s eye and the mind’s mind. With the mind’s eye, an imaginative thought, such as daydreaming about one’s friends during the COVID-19 pandemic, enters the mind with detailed, vivid mental associations and is generally low in construal, perhaps because the imagined outing is concrete (e.g., eating ice cream with friends) and grounded in a specific context (e.g., time and place). Through the mind’s mind, however, such daydreaming has greater potential to be verbalized and represented at a high level of construal, capturing the broader meaning of an event such as the impact of the pandemic on social relationships. As illustrated by this example, we suggest that many imaginative thoughts reflect a combination of the mind’s eye and the mind’s mind, which may shift in precedence and shape the unfolding of imaginative thought. Although the default-network components supporting the mind’s eye and mind’s mind are debated, we suggest that the mind’s eye is most closely associated with the medial temporal lobe (MTL) subsystem described by Andrews-Hanna et al. (2014) as well as a default-network subsystem (DN-A) described by Buckner and DiNicola (2019). Conversely, we propose that the mind’s mind is linked to the dorsal medial prefrontal cortex (dMPFC) subsystem described by Andrews-Hanna et al. and possibly another default-network subsystem (DN-B) described by Buckner and DiNicola.
The mind’s eye refers to a style of imagination in which contextually specific, concrete, and perceptually detailed thoughts arise in the form of mental associations and images. When we visualize the storyline of a good book, navigate through familiar environments, or mentally relive past events, the mind’s eye allows us to focus on the fine-grained details of our lives. Supporting our ability to retrieve and relate together items, objects, spatial locations, and other concrete forms of knowledge, the mind’s eye also affords us the flexibility to construct new ideas and mental scenarios, such as envisioning what might lie ahead, dreaming vividly, or thinking creatively. Drawing on evidence from a range of imagination tasks, we link the mind’s eye with a default-network subsystem involving the medial temporal lobe and its cortical connections, including midline parietal regions spanning the retrosplenial cortex and parts of the posterior cingulate cortex, a lateral parietal region called the inferior parietal lobe, and other cortical and subcortical regions (Andrews-Hanna et al., 2014; D’Argembeau, 2020; Sheldon et al., 2019).
Conversely, there is a rapidly growing literature highlighting that much of imagination emerges as abstract thoughts, inferences, and reflections. We propose that these more verbal features of imaginative thought are supported by a second facet of imagination, which we give the broad label the mind’s mind (Abraham & Bubic, 2015; Andrews-Hanna et al., 2014; Renoult et al., 2019). We suggest that the mind’s mind relies on a second default-network subsystem, including a dorsal midline part of the prefrontal cortex called the dorsal medial prefrontal cortex, a region at the junction of the temporal and parietal lobe called the temporoparietal junction, and anterior parts of the temporal lobe (including the temporal pole), among other medial and lateral frontal and parietal regions. Although traditionally discussed for their role in theory of mind—the process by which we can understand the thoughts and feelings of others—additional evidence indicates that these regions may also allow us to reflect on our own thoughts and emotions, retrieve abstract conceptual knowledge, comprehend narratives, and approach our episodic memories from a broader, more thematic perspective (e.g., meaning making; Andrews-Hanna et al., 2014; Gurguryan & Sheldon, 2019). The precise neural overlap among the processes of the mind’s mind marks an important area of future research.
Critically, according to our integrative framework, the mind’s mind collaboratively and dynamically interacts with the mind’s eye in many everyday contexts, including during unconstrained passive states. For example, the way in which we interpret and assign value to our memories or external world is often a balance of thoughts grounded in images and words, guiding how our experiences shape our self-identity. Consistent with this idea, many tasks grouped under the umbrella of imagination appear to activate large overlapping parts of the default network (Spreng et al., 2009). Relatedly, although many cognitive tasks target more complex forms of imagination, there is a growing appreciation in line with our framework that the default network is involved in more basic forms of cognition stretching beyond that immediately available to the senses (e.g., Konishi et al., 2015), an area that could be explored further in future research using carefully selected experimental control conditions.
Although the mind’s eye and mind’s mind are naturally interactive, recent research with neuropsychological and mental health populations has supported the distinction between them and, in the process, highlighted important clinical implications of our model. For instance, damage to the medial temporal lobe extends beyond impairments in episodic memory and episodic future thinking (Irish & Piolino, 2016) to also affect other forms of imagination relying on the mind’s eye, such as vivid atemporal scene imagery (Hassabis et al., 2007), the retrieval of autobiographical facts attached to spatiotemporal contexts (Grilli & Verfaellie, 2016), and the use of more imageable words in event narratives (Hilverman et al., 2017). In regard to the mind’s mind, individuals with semantic dementia (affecting anterior lateral temporal regions of the second subsystem) tend to exhibit better comprehension of highly imageable relative to more abstract words (Jefferies et al., 2009) and better memory for recent episodic memories, which tend to be more perceptual (Irish, 2020). Although these studies ask how a loss of aspects of the default network might affect imagination, there has been equal interest in how “overuse” of some default-network regions, which is common in psychiatric conditions involving negative repetitive thinking, can shed light on organizational features of imagination by altering their frequency and intensity (Andrews-Hanna et al., 2020). Collectively, these clinical approaches suggest that broad imbalances can occur in imagination, depending on the degree to which default-network subsystems are affected.
Our integrative framework also has implications for so-called adaptive functions of imagination. The list of proposed functions is impressive and includes facilitating our ability to solve problems with novel and creative solutions, choosing delayed over immediate rewards, thinking counterfactually (i.e., asking “What might have been?”), following through with goals, navigating the social world, and more. Mapping each of these adaptive functions to the default network will take time, although some important insights have already emerged and in ways that support our integrative framework. For example, recent research using transcranial magnetic stimulation, which can temporarily perturb cognitive functioning, suggests that both episodic future thinking and divergent creative thinking depend on aspects of the medial temporal lobe subsystem (Thakral et al., 2020). At the same time, research on creativity and other imaginative processes highlights that the default network does not operate in a vacuum but rather collaborates with other large-scale brain systems, such as those involved in executive control, emotion, and sensorimotor functions, to inspire many forms of imagination (Beaty et al., 2016; Zabelina & Andrews-Hanna, 2016).
In addition to imaginative processes that we initiate deliberately and of which we are consciously aware, a considerable amount of imagination may emerge spontaneously, as the mind wanders freely from thought to thought (Christoff et al., 2016). Relatedly, we suggest that the default network may operate largely below the surface of conscious awareness, accommodating a number of implicit cognitive processes. These spontaneous or unconscious forms of the mind’s eye and mind’s mind may also serve adaptive functions, supporting replay of neural ensembles to help stabilize memories, leading to moments of spontaneous recall of unresolved current concerns, and sparking inspiration and creativity (Christoff et al., 2016). How such processes relate to spontaneous-activity correlations is also an exciting new area of research. Below, we survey three theoretical and methodological paths forward in psychology and neuroscience’s quest to chart the landscape of imaginative thought.
Toward a Neuroscience of Everyday Imagination
Everyday imagination: the promise of naturalistic assessments of cognition
A major theory underlying imaginative thoughts is that they provide a beacon of concerns, desires, and goals that we deem personally significant, so much so that they often lead us astray from more immediate tasks at hand (Klinger, 1971). It seems logical, then, that these intimate expressions of self should be characterized in the personal contexts that shape them—our everyday lives.
Although we often assume that our laboratory tasks capture thoughts similar to experiences in the wild, recent research has painted somewhat different pictures of laboratory and naturalistically assessed imagination. For example, whereas task-unrelated thinking in the lab often predicts poor attentional control, task-unrelated thinking tends not to be as detrimental when assessed in everyday life using interspersed text-message-like surveys called ecological momentary assessments (Kane et al., 2017). Contextual influences on imagination may be particularly pronounced in older adults and clinical groups, for whom laboratory settings, experimenter demographics, and the scrutiny of performance evaluation may be perceived as more psychologically dissimilar to experiences typical of daily life. Indeed, some facets of everyday cognition that are aided by imagination, such as prospective memory (i.e., remembering to follow through on a plan) are less impaired, or even improved, in older adults compared with younger adults (Verhaeghen et al., 2012).
To be clear, we are not suggesting that researchers abandon laboratory-based studies. Rather, naturalistic assessments of imagination across a diverse set of personal contexts, aided by smartphone-based tools and other mobile devices (Fig. 2, left), can serve as a useful complement to the exquisite experimental control afforded by laboratory-based tasks. Naturalistic assessments can extend beyond answering questions about generalizability of imaginative phenomena; they can also lead to novel discoveries that can be further explored or manipulated in follow-up laboratory studies. For example, using a smartphone app called the Electronically Activated Recorder to track daily conversations, we recently showed that older age is associated with less overall sharing of autobiographical memories in daily conversations (Wank et al., 2020). Our research is now beginning to probe the cognitive and neural sources of such decline using experimental manipulations and neuroimaging in laboratory settings.
Charting new directions in the psychological science of imagination. In this article, we describe three theoretical and methodological paths forward in the psychological science of imagination. Our first path (left) leads to the naturalistic assessment of imagination, revealing the impact of personal contexts, outside of the laboratory, on how we remember, imagine the future, think creatively, mind wander, and so on. The rapid advancement of smartphone-based tools makes possible a deeper exploration of imaginative thought in the wild. Similar to how walking a trail into uncharted territory can change the feel or experience of a hike, so too can our thoughts evolve, which captures our second path forward: the dynamics of imagination (middle). Emerging stream-of-consciousness and experiencing-sampling techniques, along with advanced neuroimaging-enabling activity and connectivity to be measured dynamically, will aid navigation down this second path. For our third path (right), we highlight person-centered approaches to characterize the unique fingerprints of our thoughts at both a cognitive and a neural level. Further advancement of individual-differences approaches can guide us down this path.
Understanding the dynamics of imaginative thought
Much of what we currently know about imagination comes from paradigms that take a static approach to assessing imaginative thought. According to a recent neurocognitive taxonomy of thinking, however, dynamic properties of thought can lead to deeper understanding of disparate imaginative states, which may ultimately explain variability in cognitive and mental health traits (Christoff et al., 2016). Thoughts that flow freely and transition with ease may signify states such as dreaming, mind wandering, and creative thinking—all forms of “spontaneous thought” that emerge when constraints on cognition are relatively relaxed. Conversely, stability and persistence in thought are characteristics of deliberately constrained thinking as well as ruminative and emotionally salient thoughts that emerge outside of individuals’ deliberate control. By understanding how these static and dynamic states are expressed over longer time scales, we can gain better insight into traitlike characteristics such as creativity, personality, attentional control, and psychopathology.
Looking ahead, we call for wider use of approaches seeking to shed light on the short-term and long-term dynamics of thought (Fig. 2, middle). Our integrative framework suggests that imaginative thoughts may flow from images to words, for example, and across boundaries between types of imaginative thought. The nature of these dynamics could be revealed by assessing the duration or lability of a thought using experience-sampling techniques, in which aspects of one’s experience are sampled periodically over time (Mills et al., 2018). Relatedly, stream of consciousness or think-aloud paradigms are ideal for capturing continuous streams of thought through written or vocal responses (D’Argembeau & Mathy, 2011). In addition, new advances in functional neuroimaging permit brain activity and connectivity to be assessed dynamically and potentially in parallel to the flow of imaginative thoughts (Zabelina & Andrews-Hanna, 2016).
The promise of person-centered approaches
Despite the appreciation that our imaginative thoughts shape our self-identity, most research on imagination glosses over data at the level of the individual, favoring psychological and neural principles across groups of individuals, such as young adults compared with older adults. The final path we chart forward maintains that a person-centered approach applied to imagination can uncover new discoveries about the behavioral and neural complexity of imaginative thought.
In line with a person-centered approach, our unifying framework for imagination suggests that there are core cognitive features, some driven by the mind’s eye and others by the mind’s mind, that can lead to different styles of imagination across people. These unique mental landscapes, or cognitive fingerprints, of imagination can explain between-person variability in personality, mental health, demographic characteristics, behavioral outcomes, and more (Fig. 2, right; Andrews-Hanna et al., 2013; Kane et al., 2017). To fully evaluate this landscape, future research must move beyond sampling select features of imagination and instead adopt methods that simultaneously sample a wide array of features of imagination in an individual.
Recent research reveals that cognitive fingerprints can be accompanied by neural fingerprints—unique patterns of brain connectivity or activity underpinning cognitive variability. We suggest that a person-centered approach can map cognitive fingerprints of imaginative thought to the brain, illuminating the influence of subtle variations in the shape and location of brain structure and function. Indeed, person-focused analyses applied to the default network suggest a more intricate functional organization than previously appreciated, with interwoven “A” and “B” default-network subsystems (Buckner & DiNicola, 2019) that we suspect may map to individual differences in the mind’s eye and the mind’s mind.
In line with the personalized-medicine movement, in which treatments are not one size fits all but rather tailored to each individual, person-specific approaches could also have direct clinical relevance. For example, such approaches could inform target areas for surgery resection and electrode implantation, point to the ideal positioning of neuromodulation devices, and lead to therapeutic interventions tailored to an individual’s unique cognitive fingerprint.
Summary
The psychological science of human imagination is evolving quickly. This movement has provided clarity on the many landscapes imagination can take, from the specific and contextual characteristics of the mind’s eye to the more abstract and reflective characteristics of the mind’s mind and how the brain’s default network makes it all possible. However, there also remains much to be explored and discovered, including how real-world contexts, temporal dynamics, and individual differences shape the imaginative mind. We believe that these new paths forward and the methodological opportunities we now have for navigating these territories highlight why imagination will remain an enduring current direction in psychological science.
Recommended Reading
Andrews-Hanna, J. R., Smallwood, J., & Spreng, R. N. (2014). (See References). A comprehensive review and functional MRI (fMRI) meta-analysis that brought attention to the heterogeneous neural landscape of the default network; proposed core cognitive processes associated with imaginative, or “self-generated,” thought; and discussed clinical outcomes of optimal versus under- and overutilization of the default network.
Christoff, K., Irving, Z. C., Fox, K. C. R., Spreng, R. N., & Andrews-Hanna, J. R. (2016). (See References). A theory piece (with clinical relevance) that proposes a dynamic neuroscientific framework for understanding how thoughts emerge and unfold.
D’Argembeau, A., Cassol, H., Phillips, C., Balteau, E., Salmon, E., & Van der Linden, M. (2014). Brains creating stories of selves: The neural basis of autobiographical reasoning. Social Cognitive and Affective Neuroscience, 9(5), 646–652. https://doi.org/10.1093/scan/nst028. A clever fMRI study that provides empirical support for the role of default-network subsystems in reflective versus concrete or detailed forms of autobiographical remembering and that provides neuroscientific support for the role of the default network in imagination with the mind’s eye and the mind’s mind.
Schacter, D. L., & Addis, D. R. (2007). The cognitive neuroscience of constructive memory: Remembering the past and imagining the future. Philosophical Transactions of the Royal Society B: Biological Sciences, 362(1481), 773–786. https://doi.org/10.1098/rstb.2007.2087. An insightful, cognitive-neuroscience overview of memory and future thinking and the role of the default network that is quickly becoming a classic.
Tamir, D. I., Bricker, A. B., Dodell-Feder, D., & Mitchell, J. P. (2016). Reading fiction and reading minds: The role of simulation in the default network. Social Cognitive and Affective Neuroscience, 11(2), 215–224. https://doi.org/10.1093/scan/nsv114. Explores the contribution of the default network to vivid and detailed as well as abstract and social forms of imagination during reading, provides strong support for the mind’s eye and mind’s mind integrative neurocognitive framework, and discusses their adaptive significance.
