RESEARCH DOSSIER
Imagination and Mental Simulation
Abstract. Imagination and mental simulation permit cognition to construct and evaluate objects, actions, scenes, and possible events beyond immediately perceived conditions. This fundamental research monograph examines experimental origins, analog and propositional representational theories, spatial transformations, visual and auditory imagery, constructive episodic memory, prospective and counterfactual thought, emotion, motor rehearsal, individual variation, and risks of source-memory distortion. The analysis integrates original behavioral experiments, neuropsychological findings, psychophysical measurement, neuroimaging, and evidence-focused scholarly reviews. It distinguishes sensory vividness from task competence, neural overlap from causal necessity, and plausible simulations from accurate predictions. The synthesis characterizes imagination as a family of coordinated representational operations whose benefits depend on domain, task constraints, cognitive resources, and independently evaluated outcomes. It also identifies comparative research questions arising from human–artificial collaboration in the Artificial Era.
Keywords: imagination; mental simulation; mental imagery; mental rotation; visual imagery; episodic future thinking; constructive episodic simulation; counterfactual thinking; prospection; scene construction; aphantasia; motor imagery; cognitive neuroscience; Artificial Era
Imagination and Mental Simulation as a Scientific Object
Imagination and mental simulation concern the construction, maintenance, transformation, and evaluation of internally generated possibilities. Mental imagery denotes experiences with sensory-like properties when the corresponding external stimulus is absent; simulation more broadly denotes operations that generate a model of a situation, process, action, or possible event and explore what might occur within it. A simulation can be visual, auditory, motoric, spatial, narrative, or comparatively abstract. Its scientific identity comes from the operations and consequences that can be observed, rather than from the everyday tendency to call every conceivable thought an image. The present research object therefore encompasses transformations of represented objects, construction of imagined scenes, prospective episodes, and counterfactual changes to remembered situations. Pearson (2019) reviews the sensory dimension of human visual imagery, while Szpunar et al. (2014) distinguishes different future-directed operations.
A comprehensive account requires separate specification of content, format, construction procedure, temporal orientation, subjective experience, and behavioral function. Content concerns what is represented: a rotated object, an anticipated conversation, or an alternative outcome. Format concerns the way information is organized for processing, such as spatial relations or linguistic descriptions. Construction specifies how elements are retrieved and recombined. Temporal orientation differentiates recollection, present alternatives, and future projection. Phenomenology includes vividness, first-person perspective, affective intensity, and the sense of being situated in a scene. Function concerns prediction, rehearsal, evaluation, decision, and creative exploration. These distinctions enable direct comparisons across experimental tasks without assuming that an object image, a social scenario, and a prospective personal event are identical mechanisms.
The central research questions are which shared operations support these different capacities, how their representational resources relate to perception and memory, when imagery affects action or emotion, and which individual and contextual conditions alter simulation quality. The method of this monograph is a critical narrative synthesis of identifiable experimental studies, neuropsychological evidence, functional neuroimaging, measurement research, and domain-focused reviews. Its claims are organized by what each design can establish. Reaction-time regularities constrain process models; intervention experiments establish context-specific causal effects; neural correlations identify possible implementation systems; individual-difference associations describe covariation. These are complementary tests of a complex psychological capacity, and their convergence is more informative than any one result.
Historical Foundations of the Experimental Study of Imagery
Early scientific interest in images concerned whether inner experiences could be characterized and compared between observers. Twentieth-century cognitive experiments transformed this issue into a set of measurable problems about spatial operations, perceptual interference, and memory performance. The landmark study by Shepard & Metzler (1971) used judgments about pairs of depicted three-dimensional objects and found that response time increased approximately linearly with the angular difference between the drawings. The result made a seemingly private operation accessible through systematic variation of an observable experimental parameter. Yet the inference from a reaction-time function to a particular internal representation was always a theoretical inference, and later disputes centered on how many different mechanisms might generate the same behavioral relation.
In a complementary series, Kosslyn et al. (1978) asked participants to scan distances across images of learned displays. The dependence of response time on represented distance suggested metric constraints in visual imagination and led to more systematic investigations of image generation, inspection, and transformation. These procedures established a productive research program because stimulus geometry and task instructions could be manipulated independently. They also raised methodological questions about demand characteristics, explicit strategies, and tacit knowledge. The value of those early experiments therefore extends beyond confirmation of an image metaphor: they demonstrated how alternative explanatory assumptions could be tested against controlled measures instead of being settled by introspective description alone.
Subsequent decades widened the subject from transformations of object-like images to the construction of autobiographical scenes, possible futures, emotional events, and planned actions. Schacter et al. (2007) argued that episodic memory resources help make possible imagined futures; Hassabis et al. (2007) examined novel scene construction in patients with hippocampal amnesia. Neuroimaging studies added a distinct level of evidence about overlapping yet differentiated brain networks. Research on imagery variation later introduced objective psychophysical probes alongside reports of vividness, and Zeman et al. (2015) described lifelong absence of voluntary visual imagery in a small case series. Contemporary imagination research consequently includes both performance mechanisms and the diversity of conscious sensory experience, a development that prevents one laboratory task from serving as a complete definition.
Depictive, Propositional, and Hybrid Accounts of Representation
An enduring theoretical question asks what form an imagined scene takes in the cognitive system. Depictive accounts propose that some imagery representations preserve spatial relations in a format whose geometry supports transformations and inspections. Propositional accounts explain apparent images through structured descriptions and operations over symbolic information. The critique developed by Pylyshyn (1973) emphasized that descriptions of a mental picture do not themselves establish a picture-like internal medium. Later arguments about tacit knowledge asked whether participants reproduce familiar perceptual timing because they implicitly know how real-world perception should unfold. The dispute is scientifically consequential because a matching behavioral profile can be reproduced by different algorithms, and a participant's report cannot alone adjudicate among them.
Image-based models gain support when experimental manipulations reveal properties specific to spatial or sensory processing. Kosslyn (2005) developed the case for the functional involvement of early visual cortex during visual imagery. Other evidence is compatible with mixed representations: verbal labels, conceptual category knowledge, scene schemas, and sensory patterns may jointly determine performance. Finks et al. (1989) investigated the possibility of reinterpreting visual patterns while imagining them, an approach that asks what new information an internally represented configuration can support. Such studies are particularly valuable when the experiment requires a transformation or discovery that cannot be completed merely by repeating an instruction. They still require appropriate controls for prior recognition, verbal rehearsal, and response expectation.
Contemporary synthesis is best expressed as a set of testable representational claims at specified levels. A task may use spatially organized information without every step occurring in a pictorial buffer, and cortical overlap with perception need not mean that imagery reproduces all perceptual processing. Hybrid models allow memory-derived features, abstract goals, and sensory simulations to interact across stages. Their explanatory strength depends on specifying what changes in processing when vividness, stimulus similarity, cognitive load, and interference are experimentally altered. Dijkstra et al. (2019) reviews the evidence for shared neural representations of perception and imagery, while also making clear that overlap varies across regions and circumstances. The decisive question becomes which representation is causally necessary for each outcome.
Simulation, Prediction, and Counterfactual Possibility
A simulation constructs a structured possible course of events and permits operations upon that course. Its target may be what will happen, what could have happened, what another person may do, or what an action would feel like. This breadth requires analytic discipline. Predictive inference can occur without a consciously vivid scene, whereas perceptual imagery may occur without any intention to forecast. Prospection is an umbrella category for future-directed cognition; Szpunar et al. (2014) distinguishes simulation, prediction, intention, and planning as operations with different requirements. A prospective scene represents a possible experience; an intention marks commitment to a future course; planning arranges means under constraints; prediction evaluates likelihood. These distinctions matter for operationalizing a proposed mechanism.
Counterfactual cognition constructs alternatives to an event understood to have occurred. Kahneman & Miller (1986) described how a person's sense of normality depends partly on the alternatives that can be readily recruited, and Roese (1997) reviewed mechanisms and functions of counterfactual thinking. The alterable features of an event are rarely arbitrary. Causal salience, unusual actions, social norms, and personal goals influence which elements become mentally mutable. Upward alternatives describe a better-than-actual possibility, while downward alternatives describe a worse one. Both can influence emotional interpretation and future preparation, yet their psychological consequences depend on focus, controllability, and whether a feasible action follows from the imagined comparison.
The unifying scientific claim is conditional: different simulations may recruit shared capacities for retrieval, recombination, sequencing, and evaluation while depending on domain-specific representational resources. A future conversation requires social knowledge, language, and belief attribution; rotating a solid requires relations among spatial axes; picturing a frightening scene recruits affectively significant information. Buckner & Carroll (2007) discusses a common network implicated in self-projection across memory, future thought, and perspective taking, but a shared network is a starting hypothesis about computational commonality. Research must continue to identify which operations are shared, which processes are independently variable, and which behavioral consequences require a particular sort of simulated content.
Conceptual Boundaries and Relations to Neighboring Sciences
The object of this dossier intersects several established research areas without absorbing their full domains. Perception investigates processing of currently available sensory information, whereas imagery experiments investigate how sensory systems behave when a stimulus is recalled, instructed, or internally generated. Memory investigates encoding, retention, reconstruction, and retrieval; mental simulation examines how retrieved components are reorganized into possibilities that may never have occurred. Creativity research studies generation, selection, and evaluation of ideas and products; imagined transformations can contribute to generation, but creative value requires additional criteria. Decision research asks how possible outcomes influence choice, which is a downstream use of simulation rather than its entire mechanism.
These boundaries determine the interpretation of shared experimental paradigms. A mental rotation task yields evidence about spatial transformation, but does not directly measure imaginative originality or autobiographical future thinking. An autobiographical future interview provides insight into event construction and phenomenology, but its verbal output is shaped by memory accessibility, language, and social expectations. Affective imagery experiments measure changes in emotion or bodily response under specified instructions, yet cannot by themselves establish therapeutic effectiveness across conditions. Holmes & Mathews (2010) relates imagery to emotion, and Schacter et al. (2017) examines mechanisms and functions of episodic future thought; together these reviews show that a common vocabulary of imagination covers materially different experimental endpoints.
Interdisciplinary connections become strongest when the relation is expressed as a mechanism rather than an analogy. A model of creative generation can specify how spatial recombination enlarges candidate solutions; a model of self-regulation can show how prospective rehearsals alter preparation; a neurocognitive model can predict which lesions or disruptions impair specific constructions. Zwaan & Radvansky (1998) describes situation models during language comprehension, providing a bridge between described and internally represented events. This dossier maintains that intellectual scope while evaluating evidence by the independent validity of each task, measurement, and inference. The scientific aim is a coherent synthesis of established findings and disagreements, not a universal explanation that erases domain-specific structure.
Mental Rotation and the Geometry of Internal Transformation
The mental-rotation paradigm measures how observers compare represented objects across orientations. In Shepard & Metzler (1971), participants decided whether drawings depicted the same object under rotation or different objects, with latencies rising as angular disparity increased. The finding is consistent with a transformation process whose duration depends on the magnitude of change. It also poses an identification problem: the observed line can arise through alternative combinations of encoding, rotation-like operations, checking, and decision thresholds. Experimental interpretation depends on contrasts among mirror-image foils, orientations, axis configurations, practice, and speed-accuracy tradeoffs. Accuracy and response time should therefore be modeled together, especially when the group or individual comparison changes strategic incentives.
A geometric interpretation requires specifying the coordinate system in which transformation is carried out. Rotations in a picture plane, rotations in depth, object-centered transformations, and imagined shifts in the viewer's perspective place different demands on spatial knowledge. Participants may solve some trials by retrieving recognized views or matching characteristic features rather than continuously rotating an image. Such strategy substitution complicates claims about a unique representational algorithm. A serious model should state when angular disparity is expected to predict processing time, when familiarity changes the slope, and which alternative strategies predict nonlinearities or discontinuities. Replication of the basic monotonic relation supports a reliable task regularity while leaving detailed process architecture open to further intervention.
Mental rotation illustrates a general principle of experimental imagination: the strongest inferences come from lawful sensitivity to manipulated structure combined with controls that isolate the proposed operation. Neural activation associated with rotation can identify candidate systems, but additional tests are needed to determine whether those systems implement spatial transformation, object recognition, attention, or generic task difficulty. Similarly, a person who reports a vivid rotating object may still use feature comparison on difficult trials. The paradigm remains foundational because it offers precisely controlled problems that admit formal process models. Its scope is spatially constrained internal transformation; broader capacities such as imagining a social future must be established with their own tests and cross-domain comparisons.
Image Scanning and Metric Spatial Relations
Image-scanning tasks ask participants to inspect locations in an internally represented scene after the external display has been removed. The experiments reported by Kosslyn et al. (1978) found that longer represented distances tended to require more time for scanning, with comparisons designed to distinguish distance from the number of intervening objects. The behavioral result supports the hypothesis that imagined locations may retain metric constraints. However, a reaction-time gradient is an observation about performance, and the inference to a continuously traversed inner display remains conditional on how strongly the design controls expectancy, retrieval effort, and learned spatial relationships. A participant can know that two landmarks are far apart without maintaining a picture whose coordinates are directly inspected.
The task permits several competing predictions. If scanning depends on spatial interpolation, manipulating metric separation while holding identity and landmark count fixed should systematically affect latency. If the process primarily relies on associative retrieval, the frequency and learned strength of landmark pairings should become important predictors. If participants are explicitly instructed to imitate eye movements, demand characteristics may alter apparent scanning. These possibilities motivate variants with unfamiliar configurations, unexpected probes, nonvisual routes, secondary-task interference, and training that changes knowledge without changing geometry. The same experiment can also compare verbal and image-based instructions, provided that response demands and learning histories are sufficiently balanced.
Metric scanning bears on the structure of imagination rather than establishing that all imagined information is spatial. Prospective events can be represented through causal order, personal significance, or goals without explicit maps, whereas navigation-like construction may depend heavily on place and viewpoint. The strongest models state which spatial relations a given simulation preserves and how uncertainty in those relations affects subsequent behavior. Pylyshyn (1973) warned against equating experience of an image with a self-explanatory representational theory. The lesson for contemporary work is to retain the experimental achievement of distance-dependent performance while subjecting its interpretation to computational alternatives, instruction controls, and convergent sensory or neuropsychological measures.
Generating, Inspecting, and Reinterpreting Imagined Patterns
Imagery research distinguishes generating an internal representation from inspecting it and modifying it after generation. A participant may reconstruct a familiar scene, search for a feature, compare alternatives, or reinterpret the same configuration as a different object. Finks et al. (1989) investigated reinterpreting visual patterns in mental imagery and challenged a strong claim that mental images lack geometrical information usable for novel classifications. Such work is especially informative when a task requires an unforeseen judgment about an image rather than repetition of a known label. Yet the evidence depends on what participants knew before the prompt and how instructions guide generation, so experiments must test recognition history and exclude reinterpretation by verbal inference alone.
Transformation and reinterpretation differ in their causal architecture. A spatial rotation preserves object identity across changing orientation, while reinterpretation changes the categorical organization assigned to an apparently stable configuration. Generativity introduces a further component: elements drawn from separate experiences can be recombined into an arrangement not previously encountered. These operations may interact, but their dissociation can be tested by varying spatial complexity, semantic ambiguity, image persistence, and response alternatives. For example, an image that is easily described in words may invite a language-based strategy; an unfamiliar compound shape may force greater reliance on relation-preserving transformations. A useful experiment also establishes that the target was not inadvertently recognized before the critical trial.
Investigations of mental reinterpretation bridge visual imagination and creative cognition without providing a direct measure of creativity. Generating multiple image interpretations is a component of flexible cognition; the novelty, usefulness, and sociocultural evaluation of an outcome are separate research variables. Performance should be distinguished from the sensation of visual vividness: a participant may report a sparse image yet accurately perform a geometric transformation, and another may report great vividness without discovering a new classification. This distinction becomes essential when comparing typical observers with people reporting aphantasia. Keogh & Pearson (2018) offers psychophysical evidence on sensory imagery differences, implying that reports and task success need not be treated as interchangeable measures.
Imagery as a Modulator of Perceptual Processing
A central question is whether imagining a stimulus can alter the processing of an actual stimulus presented later. Pearson et al. (2008) studied this using the influence of visual imagery on subsequent conscious perception, providing a behavioral test of functional overlap rather than a simple comparison of introspective reports. Binocular rivalry and related competitive-perception procedures can ask whether a previously imagined orientation or pattern biases which percept becomes dominant. If the manipulation occurs before a controlled sensory display, a systematic bias suggests that imagery has changed the state of the processing system. Strong inference still requires controls for attention, expectation, decision criteria, adaptation, and voluntary response strategies.
The experimental distinction between perceptual priming and conscious imagery is important. Top-down expectation can shape perceptual decisions without forming a reportable sensory image; attention can amplify features expected in a task. A condition that requires imagining a stimulus should therefore be contrasted with conditions that ask participants to think about its label, remember its meaning, or anticipate its occurrence. Timing and stimulus strength are also critical because imagery effects may decay, saturate, or interact with weak incoming input. Evidence from psychophysics gains explanatory value when its dependent variable is objectively constrained and when subjective vividness predicts effects beyond general compliance with instructions.
The overlap is neither complete nor uniform. Dijkstra et al. (2019) reviews neuroimaging evidence suggesting related stimulus representations in visual, parietal, and frontal systems, while Pearson (2019) discusses the varied roles of sensory cortex and top-down networks in imagery. Such evidence makes it productive to ask how internally generated patterns influence perception, but an identical representational code cannot be inferred from activation of the same anatomical area alone. Cross-decoding designs, stimulus-specific patterns, and independent behavioral interference measures narrow the alternatives. The scientific target is the precise condition under which an imagined representation changes measurable perceptual sensitivity or competition and how long that influence persists.
Neural Implementation and the Limits of Overlap
Brain research has identified distributed systems recruited during visual imagery and has clarified that imagery often engages neural resources used for perception. Ganis et al. (2004) compared imagery and perception with functional magnetic resonance imaging and observed substantial overlap alongside differences in spatial distribution. The result is compatible with a model in which memory and control systems initiate sensory-pattern reconstruction. It does not prove that identical neurons carry identical information or that all imagery requires the same set of regions. Functional imaging signals represent indirect hemodynamic changes, and task comparisons are influenced by difficulty, attention, instruction, and differences between viewing a stimulus and sustaining one internally.
The role of early visual cortex remains particularly informative. Kosslyn (2005) developed arguments for its contribution to depictive visual representations, and Pearson (2019) reviews work linking anatomy and function of visual cortex to imagery strength. Other research emphasizes distributed frontal, parietal, medial temporal, and sensory contributions. The central issue is causal necessity: a region may become active while an image is maintained, yet performance could survive its disruption through compensatory or alternative pathways. Lesion studies, brain stimulation, and multivariate pattern analyses each contribute a different type of test. A strong mechanistic explanation predicts when perturbing a specific component should alter sensory vividness, spatial accuracy, imagery speed, or subsequent perception.
Neuroimaging also illustrates a general interpretative danger: reasoning backward from a region's activation to a uniquely named mental process. A hippocampal response during imagined events may reflect associative binding, episodic retrieval, scene construction, or some combination; visual-cortex engagement may track feature-specific representation without determining whether the image is conscious. Functional overlap is therefore evidence of potentially shared resources, to be interpreted alongside dissociations, interference patterns, and objective task measures. Dijkstra et al. (2019) offers a synthesis of neural similarity across visual imagery and perception, while Ganis et al. (2004) provides an original comparison. Together they justify a distributed, conditional account of neural implementation without collapsing phenomenology, computation, and hemodynamic activity into one measure.
Constructive Episodic Memory and Imagined Events
An episodic simulation presents a particular event situated in a possible place, time, and personal perspective. This kind of construction depends on memory, but it involves more than replaying an archived episode. Elements associated with different experiences can be selected, transformed, and combined into a scene whose arrangement is new. Schacter et al. (2007) develops the constructive episodic simulation hypothesis: some processes that permit adaptive remembering also support imagined future experience. This framework helps explain why neural and cognitive resources recruited during remembering frequently appear during future-event construction. The theoretical proposition is about shared components and operations, not the identity of every remembered event with every imagined future.
Studies can separate initial construction of an event from subsequent elaboration. In a functional imaging design, Addis et al. (2007) used temporal cues and distinguished the period during which participants constructed an event from the time used to enrich it with details. This is methodologically useful because the demand to locate a coherent scene can differ from adding sensory, spatial, social, and emotional properties to a scene already formed. Comparing past and future trials offers information about shared and differential recruitment, although both tasks depend on retrieval, verbal instruction, and participant compliance. Autobiographical memory has its own distortions, so a match in reported detail across conditions does not establish factual accuracy.
The scientific problem is to distinguish the construction mechanism from the product's apparent plausibility. A vivid imagined episode may contain unlikely causal transitions, while a sparse description may correspond to a well-structured feasible plan. Researchers should compare objective constraints, independent plausibility ratings, representational detail, and the specificity of temporal and spatial features. Schacter et al. (2017) reviews cognitive and neural evidence about episodic future thinking and its functions. The main synthesis is that episodic construction uses reusable memory elements, supported by relational and self-referential processes, but the mapping from remembered ingredients to imagined outcomes requires additional evaluation and monitoring.
Hippocampal Contributions to Scene Construction
The hippocampus and associated medial temporal systems are prominent in studies of episodic memory, spatial relations, and imagined scenes. Hassabis et al. (2007) examined patients with hippocampal amnesia who had difficulty imagining coherent new experiences. Their descriptions tended to lack integrated spatial scenes, supporting an important contribution of hippocampal processes to constructing imagined environments. The original evidence is compelling because it examines consequences of neurological disruption, but its small and clinically heterogeneous sample and the relationship between memory impairment and imaginative deficits require careful interpretation. A deficit may involve binding, scene layout, retrieval access, or more than one interacting operation.
Neuroimaging extends this evidence by characterizing distributed networks during remembered and imagined events. Addis et al. (2009) identified subsystems of a core network associated with constructing different classes of events, reinforcing the view that common recruitment can coexist with task-sensitive specializations. Such findings support a relational explanation: hippocampal processes may enable elements to be assembled in a coherent spatial and temporal configuration. Yet activation is compatible with several computational accounts. It remains important to distinguish relational binding from sensory imagery itself, and to compare event construction with navigation, scene perception, semantic generation, and verbal organization under equivalent task demands.
Evidence from amnesia and functional imaging should be treated as complementary rather than interchangeable. Lesions can implicate a system in a capacity, but widespread damage, compensatory strategies, and clinical differences can complicate anatomical specificity. Imaging can identify covariation within healthy participants but ordinarily does not demonstrate necessity. Schacter et al. (2007) and Schacter et al. (2017) provide broader interpretations of the constructive role of memory systems. The strongest current account holds that episodic simulation often depends on recombining distributed information into coherent scenes, with hippocampal relational operations contributing substantially. How essential particular subregions are for particular dimensions of imagination remains a distinct experimental question.
Future-Oriented Cognition, Prediction, and Planning
Prospective cognition includes imagining future experiences, estimating future events, committing to goals, and preparing sequences of action. Szpunar et al. (2014) provides an organizing taxonomy that separates simulation, prediction, intention, and planning. The distinction corrects a common ambiguity in research: participants who picture a forthcoming meeting may be simulating its experience, predicting an outcome, practicing a response, or doing several things successively. A study that treats all of these as one outcome obscures possible mechanisms. An operational account identifies the instruction, the output requested, the information available, and the criterion by which future-directed performance is evaluated.
Episodic future thinking can contribute to decisions when it makes outcomes concrete and relevant, but its effectiveness depends on accuracy and transfer. Schacter et al. (2017) reviews the prospective uses of constructed episodes. Specific future scenes can change how a person weighs choices, yet several possible mediators must be distinguished: increased emotional salience, reduced temporal distance, richer anticipated detail, and revaluation of goal outcomes. If an imagined episode alters a choice, researchers need controls for general attention, explicit persuasive instructions, and mere repetition of a future option. A plausible causal experiment manipulates the quality of event construction while measuring decisions independently rather than relying entirely on self-report.
Planning adds feasibility constraints to imagined possibility. A person may envision arriving at a destination while neglecting time costs, intermediates, and uncertainty. Process simulation can represent intermediate steps and potential obstacles, whereas outcome visualization may amplify desirability without developing an action sequence. Taylor et al. (1998) distinguishes these forms in research on goal pursuit and coping. Successful anticipation involves updating a simulation when environmental evidence changes; an attractive scenario that remains insulated from feedback can become maladaptive. Prospection research therefore requires separate outcomes for imagined detail, predicted likelihood, proposed action, actual implementation, and longer-term performance.
Counterfactual Construction, Causal Inference, and Emotion
Counterfactual simulation modifies a known or believed past event and explores the consequences of the alteration. In Kahneman & Miller (1986), the availability of nearby alternatives influences the perceived abnormality of events and associated judgments. Roese (1997) surveys the ways counterfactual thinking supports learning, attribution, and affect. One critical distinction is between undoing an outcome by changing a controllable decision and generating an alternative that differs through chance or another person's behavior. These alternatives can have different consequences for responsibility, regret, and subsequent preparation. Experimental manipulations may vary what is mutable, how easily the alternative can be generated, and whether an actionable correction follows.
The evidentiary challenge concerns the relation between subjective possibility and causal accuracy. People preferentially construct scenarios that are salient, personally important, and easy to modify. Those selection pressures can create a persuasive counterfactual that does not approximate the actual causal structure of an event. Observers may change one factor in imagination while overlooking compensating changes that would have occurred if that factor were changed in reality. A valid model should separate the generation stage, when an alternative is recruited, from the evaluation stage, when consequences are judged. Designs that manipulate norm violations or perceived controllability can help identify these stages without presuming that emotional intensity signals correctness.
Counterfactual thought connects memory reconstruction, evaluation, and prospective action. Upward counterfactuals may expose missed opportunities, while downward counterfactuals may alter relief and appraisal. Neither direction has a uniformly beneficial or harmful effect because subsequent outcomes depend on repetitive thinking, perceived agency, and available courses of action. The review by Roese (1997) provides a framework for these functions, and Kahneman & Miller (1986) grounds the role of constructed comparison standards. The scientific synthesis is that alternative-event construction is an adaptive resource under conditions where it improves causal learning or preparation, yet the same flexibility can increase regret, hindsight distortion, or overconfidence in imagined explanations.
Auditory and Multisensory Mental Imagery
Visual imagination is the best-studied sensory modality, but imagined experience also includes sound, movement, touch, and combinations of sensory features. Musical imagery offers a comparatively clear case: people can internally reproduce a melody or anticipate a passage in the absence of sound. Zatorre & Halpern (2005) reviewed converging evidence that auditory cortical regions can participate in imagined music. The finding extends the imagery-perception question beyond visual cortex, while also raising distinct difficulties of temporal sequencing, pitch representation, and auditory working memory. An imagined melody unfolds through time, so its accuracy cannot be assessed only by inspecting a static representation. Experiments may compare pitch judgments, continuation timing, interference from external tones, and recruitment of auditory brain regions.
Multisensory simulations introduce questions about binding information from separate modalities. Imagining an encounter can include the layout of a room, the sound of speech, the rhythm of a gesture, and anticipated bodily responses. These features may be coordinated without attaining equal vividness or precision. A sensory feature can also be retrieved semantically rather than re-experienced. If a participant knows that a door creaks, that knowledge need not imply hearing a specific creak in imagination. This distinction between propositional knowledge and sensory simulation mirrors representational questions in the visual domain. A rigorous experiment manipulates one modality while measuring interference, timing, or discriminability in that modality and controls demands imposed by the overall task.
Crossmodal research also cautions against universal claims about imagery strength. A person may have little visual imagery and intact auditory or spatial representations. Neural overlap with auditory perception does not establish that auditory simulation reproduces every feature of external listening. Zatorre & Halpern (2005) supports the study of sensory-specific resources; Pearson (2019) outlines a broader architecture in which higher-order processes may recruit sensory representations. Future experiments can independently assess each modality, compare how they combine during a complex imagined event, and determine when a simulated sensory component makes a unique contribution to memory, prediction, or creative construction.
Motor Imagery, Rehearsal, and Action Preparation
Motor imagery involves internally rehearsing an action without executing the corresponding overt movement. Decety (1996) reviewed evidence that imagined and performed actions recruit some overlapping neural and physiological mechanisms. The distinction between first-person kinesthetic imagery, in which a movement is experienced from within, and externally observed visual imagery matters because they can emphasize different information. Imagining lifting an object may involve predicted force and proprioception; picturing oneself from the outside may emphasize posture and spatial trajectory. Task descriptions need to specify the simulated perspective, intended movement, degree of muscle activity, and behavioral outcome rather than treating mental practice as a single uniform intervention.
Mental rehearsal can support motor preparation under certain training conditions, but the observed benefit depends on task complexity, participants' experience, fidelity of the imagined sequence, and whether imagery supplements physical practice. Activation overlap is a plausible mechanistic bridge, not an assurance of equivalent learning. The absence of overt movement makes it necessary to monitor actual muscle activity or instruction compliance when mechanism identification is important. Timing comparisons, interference with simultaneous motor demands, and performance transfer provide stronger tests than improvements in reported confidence alone. Decety (1996) offers a physiological framework for these questions and for distinguishing motor representations from mere verbal descriptions of movement.
Motor imagery also reveals why scientific categories must be operationalized rather than inferred from vividness. A participant may predict the order of movements accurately while feeling little sensory richness, and another may imagine intense bodily sensation without improving execution. Training studies should report controls, baseline skill, dose, retention interval, and physical practice exposure, because otherwise observed gains could reflect instruction, attention, or repeated testing. The relationship between imagery and action thus provides an example of the broader principle of functional transfer: an internally constructed event becomes useful when it modifies an independent measure of behavior under conditions where competing explanations have been constrained.
Emotional Simulation, Threat, and Clinical Relevance
Imagined events can evoke powerful emotional and physiological responses. Holmes & Mathews (2010) reviewed evidence that emotional mental imagery recruits pathways associated with sensory processing, affective memory, and emotion elicitation, while Ji et al. (2016) examined imagery as simulation of reality in fear and related experiences. An imagined scene may affect bodily arousal or motivation more strongly than an abstract verbal description, but this difference depends on task, individual, and stimulus. Emotional intensity is an outcome that needs measurement rather than an inherent attribute of every image. Studies may record reported emotion, autonomic reactivity, attention, or behavioral avoidance, with careful attention to how these indicators relate.
One mechanistically informative experiment is Reddan et al. (2018), which investigated imagined exposure to a conditioned threatening stimulus and measured subsequent neural and physiological threat responses. The findings suggest that deliberately imagined experiences can participate in learning processes that alter responses to real cues. The study supports a carefully bounded causal claim about a specific laboratory manipulation and conditioned-threat setting. Its results should not be generalized automatically to every anxiety presentation or therapeutic protocol. Clinical applications depend on diagnosis, treatment fidelity, adverse effects, individual histories, and research designs specifically testing patients and meaningful health outcomes.
Emotional imagery has a dual relevance: the same construction capacity can be recruited in deliberate rehearsal and can contribute to persistent distress when emotionally charged events intrude or recur. Ji et al. (2016) emphasizes that sensory simulation may make an internally generated event feel consequential in ways that differ from verbal thinking. The critical scientific challenge is to identify which elements of an image produce the observed response: sensory vividness, autobiographical relevance, expectancy, threatening content, or repetition. Work on emotional simulation is strongest when it distinguishes a demonstrable mechanism in a controlled experiment from clinical efficacy and when it recognizes that individual intervention decisions belong to qualified professional care.
Aphantasia, Vividness, and Dissociations in Ability
People differ markedly in whether they experience voluntary visual images and how clearly they experience them. Marks (1973) introduced influential procedures for assessing vividness of visual imagery, and Zeman et al. (2015) described respondents reporting a lifelong absence of the familiar sensory experience. The term aphantasia denotes a reported or measured absence or substantial reduction of voluntary visual imagery, with definitions varying across instruments and studies. Its existence focuses attention on a fundamental distinction: conscious sensory experience, information stored about an object, and accuracy on a cognitive task are separable variables. A person can possess spatial knowledge while reporting little visual phenomenology.
Psychophysical work provides an important supplement to questionnaires. Keogh & Pearson (2018) used measures of sensory imagery in self-identified individuals with aphantasia, showing that imagery-related differences can appear in performance assays as well as reported vividness. Wicken et al. (2021) examined physiological responses while people read and imagined frightening stories; reduced skin-conductance change in the aphantasia group provided evidence that visual imagery can influence emotional simulation. Such findings require attention to selection, sample composition, modality, instruction compliance, and the fact that autonomic outcomes are influenced by multiple factors. Group contrasts do not imply a single deficit shared by everyone who identifies with the label.
Measurement should map a multidimensional profile rather than place everyone on a single scale. Voluntary vividness, involuntary imagery, visual working memory, spatial transformation, autobiographical scene richness, and auditory imagery may dissociate. Pearson (2019) reviews visual imagery variation, including especially vivid experience, while Zeman et al. (2015) demonstrates the need to attend to a lesser-known endpoint of that variation. Researchers need representative sampling beyond self-selecting communities and psychometric validation of their definitions. The result is a more precise account of imagination as a family of related capacities in which subjective imagery strength provides one informative indicator, but neither the full phenomenon nor an automatic ranking of intellectual competence.
Memory Distortion and the Costs of Imaginative Fluency
The constructive nature of imagination creates opportunities for adaptive planning and also for errors in source monitoring. When internally generated details become familiar, people may later misattribute their origin to perception or autobiographical experience. Garry et al. (1996) reported that imagining childhood events could increase confidence that such events had occurred, a finding commonly termed imagination inflation. This experimental demonstration concerns confidence under a particular task and should not be conflated with proof that every imagined episode produces a durable false autobiographical memory. Research must distinguish familiarity, belief, recollective detail, and factual accuracy, which can change independently.
Source monitoring requires discriminating whether content came from an actual event, a description, inference, dream, or deliberate simulation. An imagined scene can inherit realistic details from genuine memories while containing an invented central event. Repetition and social suggestion may increase familiarity and make attribution more difficult. The original experiments of Garry et al. (1996) motivate careful assessment of baseline beliefs and subsequent change, including controls that distinguish imagination from repeated questioning or exposure. Stronger causal tests use random assignment and delayed follow-up, while ethical restrictions are important when memory-related manipulations concern emotionally significant personal events.
These findings complicate any simple celebration of imaginative vividness as a universally beneficial resource. Highly accessible simulations can support rehearsal while producing confidence that exceeds evidence. This applies to forecasts, counterfactual explanations, and autobiographical accounts alike. Roese (1997) describes the functions of counterfactual thinking, and Schacter et al. (2007) places constructive operations within a memory system that is flexible yet fallible. Effective simulation requires evaluation of the simulation's source and plausibility against independently available constraints. A vivid possibility may be psychologically powerful and epistemically uncertain at the same time; an adequate theory predicts both consequences and the conditions that separate them.
Operational Definitions and Measurement Architecture
Imagination research uses a heterogeneous set of instruments, so construct validity begins with specifying the variable each instrument measures. Visual-vividness questionnaires measure introspective judgments; mental rotation measures performance under spatially varying demands; binocular-rivalry priming estimates the sensory consequences of prior imagery; event-description tasks measure reportable features of internally constructed scenes; functional imaging measures condition-related hemodynamic signals. These outputs are informative because they sample different dimensions. They cannot be combined into one index without a theoretical measurement model showing that they reflect a shared latent construct. Marks (1973) established an influential vividness measure, while Pearson et al. (2008) offers a behavioral alternative with a different error structure.
Experimental design should distinguish manipulation fidelity from dependent-variable change. In an imagery instruction condition, researchers should verify whether participants understood the prompt, how long they maintained a representation, whether they used covert language or other strategies, and whether condition differences remain after controlling task demand. A prospective simulation study needs definitions of event plausibility, episodic specificity, temporal distance, and subsequent behavior. Spatial studies need response-time distributions, accuracy, and explicit geometric constraints. Neuroimaging requires principled selection of contrasts and correction for multiple comparisons. Addis et al. (2007) exemplifies a distinction between event construction and elaboration, and Dijkstra et al. (2019) shows how different neural comparisons lead to different conclusions.
Cross-method triangulation is particularly important where phenomenology is central. A vividness score can be affected by participants' interpretations of questionnaire terms, while imagery strength in psychophysics depends on stimulus and timing. Keogh & Pearson (2018) provides a way to compare subjective and objective measures in extreme imagery variation. Such comparisons should not presume that either scale alone provides an error-free ground truth. Researchers can predefine a family of competing models, collect multiple indicators, and examine which predictions generalize across tasks and participants. Reliable scientific measurement requires evidence of test-retest consistency, sensitivity to relevant change, discriminant validity, and freedom from preventable instruction confounds.
Evaluating Causation, Replication, and Generalizability
The evidence base contains experiments, lesion studies, functional imaging, individual-difference comparisons, and integrative reviews, each with characteristic strengths. Randomized manipulations of imagined content or practice can support causal claims about short-term outcomes under the tested conditions. Case studies of neurological injury supply evidence about potential necessity but may be affected by coincident impairments. Cross-sectional imagery surveys identify covariation but cannot alone establish developmental direction. Imaging overlap is correlational unless combined with interventions that specifically perturb a component. The scientifically appropriate claim should never exceed the causal resolution of the design. Hassabis et al. (2007), Pearson et al. (2008), and Reddan et al. (2018) exemplify three distinct research designs and inferential scopes.
Replication should establish both whether a result recurs and whether it generalizes beyond the original stimulus and procedure. In rotation experiments, a robust angular-disparity function may still vary with object familiarity or instructions; in affective imagery, a physiological result may depend on the class of threat, sample characteristics, and exposure sequence. Pre-registration, adequately powered independent replications, openly specified exclusions, and alternative analytic models improve interpretability. A pooled effect can obscure opposing results across meaningful subgroups. Evidence review must therefore ask whether measured effects concern conscious experience, task strategy, or subsequent behavior, and whether proposed mediators were manipulated directly.
Generalizability also depends on cultural and technological conditions. Autobiographical event construction uses schemas learned in particular contexts; verbal prompts can function differently across languages; access to visualization tools can change strategies for spatial reasoning. Training and assessment tools should be adapted carefully rather than assumed to measure the same operation across groups. Szpunar et al. (2014) highlights a conceptual taxonomy that facilitates sharper comparisons, while Pearson (2019) integrates multiple levels of visual-imagery evidence. In the present synthesis, a strong conclusion is one supported by multiple independent observations that agree on a specified mechanism, not simply by a large number of papers sharing the word imagination.
Goal Pursuit, Mental Contrasting, and Practical Transfer
Imagination can regulate action when constructed possibilities organize preparatory behavior. Taylor et al. (1998) reviewed experimental work indicating that simulating the process of reaching a goal can improve progress under some circumstances, whereas repeatedly fantasizing about successful completion may not provide the same benefit. The important distinction is what the imagined sequence contains. A process simulation includes actions, obstacles, and coping responses, while an idealized outcome can increase desirability without changing implementation. Research should independently measure task completion, delayed performance, and adaptation to unexpected obstacles rather than treating immediate motivational confidence as the final criterion.
Mental contrasting provides a related mechanism by juxtaposing a desired future with current reality and its constraints. Oettingen (2012) reviewed fantasy-realization theory and evidence that this procedure can facilitate commitment when prospects are favorable and disengagement when they are poor. The explanatory target includes the interaction of desired outcome, expected feasibility, and the identification of obstacles. A scientifically informative intervention specifies how the future was represented, what reality information was supplied, and what behavior changed afterwards. Positive outcome imagination, process rehearsal, and contrasting are distinct experimental conditions even when they all involve future-oriented thought.
Practical transfer remains conditional on the match between the simulated scenario and the environment in which action occurs. A detailed rehearsal can be useful when the expected obstacles are stable and recognizable; it may fail when conditions change or when imagined confidence substitutes for learning from feedback. Taylor et al. (1998) and Oettingen (2012) support an account in which simulated futures contribute to self-regulation through specified cognitive and motivational operations. Evidence is most informative when actual choices, persistence, and performance are measured separately from imagined vividness and when the research design can distinguish rehearsal effects from additional instruction, attention, or expectation.
Creativity, Generativity, and Constructed Alternatives
Creative exploration involves producing possibilities, transforming representations, and evaluating candidates. Mental simulation can contribute to the generation stage by allowing combinations that are physically absent or not yet realized. Finks et al. (1989) provides an experimental illustration of reinterpretation and representational discovery in imagery. Such work is relevant to creativity because novel configurations may afford interpretations that were unavailable before transformation. Nevertheless, generation of an internally surprising interpretation is only one contributor to creativity; task-specific usefulness, originality relative to a reference population, and external evaluation require additional evidence. The research object here is the constructive operation rather than a comprehensive theory of artistic achievement.
Simulated possibilities can also support search through alternative designs or courses of action. A participant may consider how an imagined structure would appear from another angle or how a proposed procedure might unfold under changed constraints. The value of simulation depends on the fidelity of the represented constraints and the efficiency of evaluating alternatives. Shepard & Metzler (1971) demonstrates a lawfully measurable spatial transformation, whereas Schacter et al. (2017) addresses assembling and using future episodes. Their comparison suggests a hierarchy of operations, from domain-constrained transformation to integrated scene construction. Whether a single generative mechanism explains both remains an empirical question requiring cross-task dissociations and mechanistically precise computational models.
Creative imagination also illustrates why internal simulation needs external validation. A plausible mental design can fail under physical constraints, social expectations, or material behavior not represented by the thinker. Evaluative feedback can revise both the candidate and the generator's future strategies. The scientific relation to aesthetic psychology is therefore clear and limited: imagery and simulation contribute to construction, exploration, and anticipation, while taste, evaluation, cultural recognition, and product assessment require dedicated research designs. Finks et al. (1989) and Pearson (2019) provide specific evidence about internally generated visual content; broader claims about artistic value require independent evidence beyond the present dossier.
Artificial Era and the Study of Simulative Systems
The Artificial Era expands the methodological setting in which simulations can be investigated by introducing systems that construct and transform descriptions, spatial states, plans, and prospective narratives. In this context, a simulation should be identified by an explicit representational state, an update or transformation procedure, and observable performance under tests. A computational system may generate a coherent description of an imagined event without evidence of sensory phenomenology comparable to that reported by humans. Conversely, behavioral competence in a simulation task can be assessed directly through accuracy, generalization, constraint satisfaction, and error correction. These distinctions preserve scientific testability while allowing comparison across biological and artificial implementations.
Human–artificial collaboration creates further questions about how internally generated scenarios interact with externally presented alternatives. A person may use a generative system to propose possible plans, visualize consequences, compare timelines, or articulate counterfactual questions. The psychological outcomes can include cognitive offloading, increased access to alternatives, altered confidence, or anchoring to fluent but unsupported scenarios. The existing evidence on human imagery provides relevant methodological principles rather than immediate equivalence claims. Taylor et al. (1998) demonstrates why process and outcome simulations should be separated; Garry et al. (1996) illustrates why familiarity and imagined fluency can affect belief. Both principles motivate experiments that compare unaided human simulation with aided construction under matched information constraints.
A research program for Artificial systems should predefine the simulated domain, control the available inputs, specify the intended output, and evaluate generated possibilities against independent evidence. It should distinguish human experience reports, machine behavior, and theoretical interpretation, with measured performance determining which claims are empirically warranted. The phrase Artificial Sapiens names an authorial theoretical identity within the wider Artificial Era, but it does not replace experimental evidence about subjective experience or the architecture of particular systems. The substantive contribution of the present dossier is to preserve precise human-psychological mechanisms while identifying the kinds of comparative questions that new nonbiological technologies make experimentally accessible.
Open Questions and Priorities for Future Research
Several unresolved problems define the frontier of imagination research. First, the relation among vividness, representational fidelity, and performance remains incompletely specified. Individuals may report markedly different sensory experiences while achieving similar scores on selected tasks. Keogh & Pearson (2018) and Wicken et al. (2021) provide evidence that phenomenology can also correspond to measurable sensory or physiological differences. Future work should map these dissociations across multiple modalities, use preregistered measures, and follow individuals over time. Second, the neural necessity of particular sensory and hippocampal regions depends on task design, compensatory resources, and the difference between creating a scene and elaborating it.
Third, more precise computational models are needed to link behavior to representation. The original imagery debate raised the problem of underdetermination: behavioral slopes or vivid reports can be compatible with distinct representations. Pylyshyn (1973) and Kosslyn (2005) exemplify competing emphases; Dijkstra et al. (2019) supplies newer neural constraints without dissolving every ambiguity. Model comparison should be predictive and intervention-sensitive, with the goal of identifying which components fail or change under specific perturbations. Fourth, investigators need clearer bridges between low-level sensory tasks and complex personal simulations. A measure sensitive to imagined grating orientation should not automatically be taken as an index of future planning or social imagination.
Fifth, stronger evidence is required for practical and clinical generalization. Some laboratory manipulations of emotion and goal rehearsal produce measurable effects, but intervention protocols, participant histories, longer follow-up, and adverse outcomes alter the scientific meaning of those results. Reddan et al. (2018), Holmes & Mathews (2010), and Oettingen (2012) support distinct lines of investigation whose eventual convergence must be demonstrated. Finally, comparative studies involving new digital tools should separate human-generated from externally supplied possibilities and test whether assistance improves the calibration of predictions. The enduring priority is to explain what an internally generated possibility does, how it is constructed, and how its reliability is assessed against the world.
Conclusion
Imagination and mental simulation form an interconnected family of processes that permit cognition to operate upon absent objects, alternative situations, and possible events. The history of the field demonstrates a progression from spatially controlled laboratory experiments to studies of distributed neural systems, episodic reconstruction, emotion, and future-oriented behavior. The mental rotation findings of Shepard & Metzler (1971), image-scanning experiments of Kosslyn et al. (1978), prospective-memory synthesis of Schacter et al. (2007), and psychophysical imagery work of Pearson et al. (2008) each make different facets of an otherwise elusive capacity experimentally tractable. Their collective significance lies in the possibility of specifying operations and consequences, rather than treating imagination as a single undifferentiated faculty.
The evidence supports several broad propositions. Spatial representations can constrain internal transformations; sensory imagery can interact with perceptual processing; memory systems provide components for constructing imagined events; hippocampal and distributed cortical networks contribute to aspects of scene construction; counterfactual alternatives influence appraisal; and prospective simulations can support preparation when their content connects to actionable constraints. These propositions remain appropriately delimited by their measurement methods. Dijkstra et al. (2019), Hassabis et al. (2007), Roese (1997), and Taylor et al. (1998) jointly show that neural, phenomenological, and behavioral evidence must be integrated without assuming identity across levels. A vivid image is neither a guaranteed accurate prediction nor a necessary condition for every kind of complex thought.
The most durable scientific synthesis is a relational one: generation, maintenance, transformation, recombination, affective valuation, and evaluation are distinguishable operations whose coordination enables imagined possibilities to influence memory, choice, and action. Their exact combination depends on task, modality, individual experience, and environment. Future research must improve measurement invariance, causal identification, replication, and tests of cross-domain generality. The Artificial Era provides new settings for examining simulated alternatives and human–artificial cooperation while preserving the distinction between computational output and reported experience. Imagination remains a fundamental research object because it gives cognition measurable access to possibilities beyond immediate stimulation, together with corresponding responsibilities for verification.
References
- Addis, D. R., Wong, A. T., & Schacter, D. L. (2007). Remembering the past and imagining the future: Common and distinct neural substrates during event construction and elaboration. Neuropsychologia, 45(7), 1363–1377. https://doi.org/10.1016/j.neuropsychologia.2006.10.016
- Addis, D. R., Pan, L., Vu, M.-A., Laiser, N., & Schacter, D. L. (2009). Constructive episodic simulation of the future and the past: Distinct subsystems of a core brain network mediate imagining and remembering. Neuropsychologia, 47(11), 2222–2238. https://doi.org/10.1016/j.neuropsychologia.2008.10.026
- Buckner, R. L., & Carroll, D. C. (2007). Self-projection and the brain. Trends in Cognitive Sciences, 11(2), 49–57. https://doi.org/10.1016/j.tics.2006.11.004
- Decety, J. (1996). The neurophysiological basis of motor imagery. Behavioural Brain Research, 77(1–2), 45–52. https://doi.org/10.1016/0166-4328(95)00225-1
- Dijkstra, N., Bosch, S. E., & van Gerven, M. A. J. (2019). Shared neural mechanisms of visual perception and imagery. Trends in Cognitive Sciences, 23(5), 423–434. https://doi.org/10.1016/j.tics.2019.02.004
- Finks, R. A., Pinker, S., & Farah, M. J. (1989). Reinterpreting visual patterns in mental imagery. Cognitive Science, 13(1), 51–78. https://doi.org/10.1207/s15516709cog1301_2
- Ganis, G., Thompson, W. L., & Kosslyn, S. M. (2004). Brain areas underlying visual mental imagery and visual perception: An fMRI study. Cognitive Brain Research, 20(2), 226–241. https://doi.org/10.1016/j.cogbrainres.2004.02.012
- Garry, M., Manning, C. G., Loftus, E. F., & Sherman, S. J. (1996). Imagination inflation: Imagining a childhood event inflates confidence that it occurred. Psychonomic Bulletin & Review, 3(2), 208–214. https://doi.org/10.3758/BF03212420
- Hassabis, D., Kumaran, D., Vann, S. D., & Maguire, E. A. (2007). Patients with hippocampal amnesia cannot imagine new experiences. Proceedings of the National Academy of Sciences, 104(5), 1726–1731. https://doi.org/10.1073/pnas.0610561104
- Holmes, E. A., & Mathews, A. (2010). Mental imagery in emotion and emotional disorders. Clinical Psychology Review, 30(3), 349–362. https://doi.org/10.1016/j.cpr.2010.01.001
- Ji, J. L., Burnett Heyes, S., MacLeod, C., & Holmes, E. A. (2016). Emotional mental imagery as simulation of reality: Fear and beyond—A tribute to Peter Lang. Behavior Therapy, 47(5), 702–719. https://doi.org/10.1016/j.beth.2015.11.004
- Kahneman, D., & Miller, D. T. (1986). Norm theory: Comparing reality to its alternatives. Psychological Review, 93(2), 136–153. https://doi.org/10.1037/0033-295X.93.2.136
- Keogh, R., & Pearson, J. (2018). The blind mind: No sensory visual imagery in aphantasia. Cortex, 105, 53–60. https://doi.org/10.1016/j.cortex.2017.10.012
- Kosslyn, S. M., Ball, T. M., & Reiser, B. J. (1978). Visual images preserve metric spatial information: Evidence from studies of image scanning. Journal of Experimental Psychology: Human Perception and Performance, 4(1), 47–60. https://doi.org/10.1037/0096-1523.4.1.47
- Kosslyn, S. M. (2005). Mental images and the brain. Cognitive Neuropsychology, 22(3–4), 333–347. https://doi.org/10.1080/02643290442000130
- Marks, D. F. (1973). Visual imagery differences in the recall of pictures. British Journal of Psychology, 64(1), 17–24. https://doi.org/10.1111/j.2044-8295.1973.tb01322.x
- Oettingen, G. (2012). Future thought and behaviour change. European Review of Social Psychology, 23(1), 1–63. https://doi.org/10.1080/10463283.2011.643698
- Pearson, J., Clifford, C. W. G., & Tong, F. (2008). The functional impact of mental imagery on conscious perception. Current Biology, 18(13), 982–986. https://doi.org/10.1016/j.cub.2008.05.048
- Pearson, J. (2019). The human imagination: The cognitive neuroscience of visual mental imagery. Nature Reviews Neuroscience, 20, 624–634. https://doi.org/10.1038/s41583-019-0202-9
- Pylyshyn, Z. W. (1973). What the mind's eye tells the mind's brain: A critique of mental imagery. Psychological Bulletin, 80(1), 1–24. https://doi.org/10.1037/h0034650
- Reddan, M. C., Wager, T. D., & Schiller, D. (2018). Attenuating neural threat expression with imagination. Neuron, 100(4), 994–1005.e4. https://doi.org/10.1016/j.neuron.2018.10.047
- Roese, N. J. (1997). Counterfactual thinking. Psychological Bulletin, 121(1), 133–148. https://doi.org/10.1037/0033-2909.121.1.133
- Schacter, D. L., Addis, D. R., & Buckner, R. L. (2007). Remembering the past to imagine the future: The prospective brain. Nature Reviews Neuroscience, 8, 657–661. https://doi.org/10.1038/nrn2213
- Schacter, D. L., Benoit, R. G., & Szpunar, K. K. (2017). Episodic future thinking: Mechanisms and functions. Current Opinion in Behavioral Sciences, 17, 41–50. https://doi.org/10.1016/j.cobeha.2017.06.002
- Shepard, R. N., & Metzler, J. (1971). Mental rotation of three-dimensional objects. Science, 171(3972), 701–703. https://doi.org/10.1126/science.171.3972.701
- Szpunar, K. K., Spreng, R. N., & Schacter, D. L. (2014). A taxonomy of prospection: Introducing an organizational framework for future-oriented cognition. Proceedings of the National Academy of Sciences, 111(52), 18414–18421. https://doi.org/10.1073/pnas.1417144111
- Taylor, S. E., Pham, L. B., Rivkin, I. D., & Armor, D. A. (1998). Harnessing the imagination: Mental simulation, self-regulation, and coping. American Psychologist, 53(4), 429–439. https://doi.org/10.1037/0003-066X.53.4.429
- Wicken, M., Keogh, R., & Pearson, J. (2021). The critical role of mental imagery in human emotion: Insights from fear-based imagery and aphantasia. Proceedings of the Royal Society B: Biological Sciences, 288, 20210267. https://doi.org/10.1098/rspb.2021.0267
- Zatorre, R. J., & Halpern, A. R. (2005). Mental concerts: Musical imagery and auditory cortex. Neuron, 47(1), 9–12. https://doi.org/10.1016/j.neuron.2005.06.013
- Zeman, A., Dewar, M., & Della Sala, S. (2015). Lives without imagery—Congenital aphantasia. Cortex, 73, 378–380. https://doi.org/10.1016/j.cortex.2015.05.019
- Zwaan, R. A., & Radvansky, G. A. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123(2), 162–185. https://doi.org/10.1037/0033-2909.123.2.162
Social Perspectives and Situation Models
Imagined social interactions recruit knowledge about people, norms, intentions, and sequences of action. Buckner & Carroll (2007) proposed a self-projection framework linking episodic memory, prospection, and perspective taking through partly overlapping systems. The hypothesis is valuable because all three tasks require departure from immediate sensory conditions. Yet the meaning of another person's perspective cannot be reduced to the personal viewpoint used in autobiographical imagery. Social simulation adds uncertainty about beliefs, preferences, and the other person's access to information. Experiments must therefore distinguish scene construction from belief reasoning, narrative comprehension, familiarity, and generic difficulty.
Situation models provide a related perspective on comprehension. Zwaan & Radvansky (1998) reviewed research showing that readers construct integrated representations of described events, including temporal, spatial, causal, and intentional relations. A sentence-by-sentence update to a situation model differs from free autobiographical imagination, but both require maintaining structured relations among elements in a scene. This comparison motivates tasks in which participants read, hear, or observe an event and later answer probes requiring information about a changed situation. The data may reveal which relations remain available and which updates impose processing costs. Such measures are behaviorally informative even when participants do not report an especially vivid inner image.
Social simulation is especially susceptible to overgeneralization when a persuasive imagined narrative is mistaken for evidence about another person's actual motives. One can internally rehearse dialogue, anticipate a response, or compare alternative social choices while relying on incomplete knowledge. The quality of the simulation should be evaluated against later observations or independent information, rather than vividness or confidence. The theoretical proximity of self-projection and situation modeling suggests common operations of structured event construction, yet domain-specific resources remain crucial. Buckner & Carroll (2007) and Zwaan & Radvansky (1998) together support research that examines shared event representations alongside distinct requirements for mental-state inference, social feedback, and contextual revision.