I. Introduction

This paper works to explain the social patterning of a particular sort of unusual experience: the sighting of an unidentified flying object (UFO). Typical research on UFOs has presumed that firsthand UFO experiences are symptomatic of psychosocial deviance. Yet such individual-level explanations have not done a good job explaining empirical variation in sightings. A different sort of sociological approach can do better. Specifically, as I argue here, UFO sightings are the result of a particular form of environmental infrastructure that structures people’s vision in such a way that UFOs are seen. UFO sightings are best understood as an ecological outcome rather than as a deviant, cognitive, or dispositional outcome.

In making this argument, I advance sociological theories of place by bringing them into conversation with the sociology of the senses and the sociology of practice, through the concept of “affordances.” Over two decades ago, Gieryn (2000) penned an Annual Review article advocating for a “space for place” in sociology, lamenting that “sociologists have given the appearance of not being interested in place” (464) in order to protect the autonomy of their preferred “social and cultural variables” and modes of explanation. As he suggests, nearly all phenomena of interest to sociologists unfold in specific places. The spatialized nature of human affairs is reflected in the great many sociological studies whose units of analysis are areal, from neighborhoods to nation-states. Nevertheless, even when observations are spatial in nature, places within sociological research designs are often treated as containers of abstract sociodemographic and economic variation. While suitable for many research questions, this treatment of place overlooks its phenomenological and pragmatic dimensions: the sensual experience of place and the moment-to-moment “texture” (de la Fuente 2019) of emplaced practices and routines.

In a seemingly unrelated development, cultural sociologists have taken a recent interest in sensory experience, building on a legacy of theorizing the body as socially and historically contingent. Validating Marx’s (1975, 353) assertion that “the cultivation of the five senses is the work of all previous history,” these studies have demonstrated that our sense perceptions—of perfumes (Cerulo 2018; Rawlings and Childress 2021) or songs (Lembo and Martin 2022), for example—are interpreted through cultural schemas and “somatic work” (Waskul and Vannini 2008), such that most people intuitively associate a fragrance, for example, with particular social categories and contexts. Sense perceptions can systematically vary as a function of individual identities (Lembo 2020) and interactional contexts (Klett 2014; Rose-Greenland 2016), even mediating boundaries between social groups (Wohl 2015). In brief, this growing subfield within cultural sociology has treated sense perceptions as (at least indirectly) observable phenomenological artifacts that reflect biographical regimes of socialization, demonstrating that even our most seemingly unmediated and direct sensory experiences are social in nature.

In this paper, I explain the social patterning of UFOs by highlighting the entanglement of local sociomaterial infrastructures, habitual practices, and the ongoing generation of distinct and emplaced sensory experiences. This effort builds upon recent work that illuminates the overlooked ways in which culture and habit are both fundamentally spatial (Richer 2015) and dependent upon interaction with nonhuman, material aspects of social environments (Cerulo 2009; McDonnell 2010; Jerolmack and Tavory 2014; Wood et al. 2018; Adler Jr. et al. 2022).

I work to advance recent sociological engagement with affordance theory (Gibson 2015; Davis 2020). Originating in the psychology of visual perception and adapted to a wide range of academic contexts, affordance theory anticipated contemporary interest in the material and environmental prerequisites for action, while stressing the dual relationship that binds the pursuit of locally embedded actions to the generation of sensory experiences. In short: perception literally occurs in and through our actions, and what we can do in any environment depends upon our embodied skills that confront a material environment loaded with possibilities for action. Those possibilities—and therefore the sensory experiences they enable and entail—are best understood as a latent property of the entire organism-environment system rather than of individual and isolable cognitive processes.

Extending affordance theory’s paradigmatic concern with the action-perception-environment nexus to sociological theories of practice and place, I describe how the unique “feel” of a place derives in large part from its distinctive assemblage of affordances—in the modern era, historically produced and remade by flows of capital—that encourage certain patterns of action and perception but not others. I argue that this approach offers new ways to make sense of the social distribution of experiential phenomena that social scientists have at times divorced from the concrete and sensual moments that generate them.

I begin by introducing affordances in more detail to identify their strength for the question at hand. After demonstrating that the concept of affordances, sociologically enhanced, parsimoniously links places, sensory experiences, and practices, I apply this framework to a particular kind of place and a particular kind of sensory experience: the rural “recreational” county and the UFO sighting.

Rural recreation counties are the result of shifts in global capitalism in the late twentieth century that dramatically reoriented the affordance structures of certain rural areas away from traditional rural industry and toward outdoor recreation and leisure. These changes in affordances entail a concomitant transformation of localized sensuality, as sociomaterial infrastructures—from domestic architecture to conservationist reforms—work to suffuse even the most mundane of regular practices with the sensual and aesthetic consumption of nature. If the practical routines and affordance structures associated with traditional rural occupations—farming, mining, ranching—condition attention downward to the dense materiality of the earth, those constitutive of contemporary rural recreation direct attention toward natural environments valued for their aesthetic and leisurely potential rather than for their productive potential: people look up toward the heavens (Figure 1).

Figure 1

Left: The Gleaners, Jean-François Millet (1857), an iconic representation of nineteenth-century rural labor. Right: A home in Aspen, Colorado with floor-to-ceiling windows.

Empirically, I analyze a seemingly odd dataset: a large record of UFO sightings in the rural United States from 2000 to 2020. Through this analysis, I demonstrate that thinking with affordances can illuminate pragmatic and phenomenological regularities underlying even normatively inexplicable or irrational experiences.

I do not propose this affordance-based approach as an overarching metaphysics or methodological panacea, nor do I choose exotic data for provocative purposes. What one could do with UFO sightings, one could also do with any category of experience resulting from probabilistic, sensual encounters with social environments, a point I return to in the discussion. I merely suggest this way of thinking to help focus research designs, especially quantitative ones, on a particular bandwidth of variation related to the sensory experience of different places. UFO sightings are particularly useful data to this end, because they are inherently visual and environmentally situated experiences whose relationships to vision and to human ecology have been almost entirely ignored by the social scientific imagination. The variation in UFO sighting rates that I model suggests that there is value in systematically returning to the sensory foundation of certain social experiences when we ask why some people have those experiences and others do not.

II. Affordances for sociology

The psychologist James J. Gibson (2015) introduced the term affordance in his theory of visual perception that radically challenged the cognitivist orthodoxy of psychology at the time. A strongly cognitive approach to vision imagines the eyes transmitting unordered signals to the brain, at which point cognitive structures and processes transform those signals into the recognizable and actionable components of experience. From this perspective, vision is the experience of a “mental model” of the environment rather than of the environment itself, a model that is constructed prior to and enables subsequent embodied actions. The resulting framework posits a clear priority of mind over body, reinforcing the notion that we can only access the physical world around us through the impassable intermediary of cognitive representations.

Pushing back against this “brain-in-a-jar” image, Gibson theorized visual perceptions as emerging from the holistic interface of active organisms and environmental affordances—those material aspects and elements of an immediate environment that afford (or constrain) opportunities for action and perception—leading him to describe his theory as “ecological.” The sorts of movement and action afforded by a particular element of the environment depend both on its material properties and the interacting organism’s physiology: its size, whether it can fly or walk up walls, and so on. Through the movements and actions afforded by an environment, we vary our experience of what he called the “ambient optic array”—the structure of light relative to a point of observation—and at the same time discover what is invariant about it. To visually perceive is to extract these invariants which provide the information that specifies our local environment (Gibson 2015, 58–84). Perception, then, is not a mental prerequisite for action, but rather emerges through direct engagement with environmental affordances—an achievement of the entire ecological system rather than of an isolated mind—anticipating similar ideas in embodied cognition (e.g., Clark 1999; Noë 2004) and echoing philosophical traditions of direct realism (Gibson 1976, 235) and phenomenology (e.g., Merleau-Ponty 1962, 1964).

Gibson positioned affordance theory at the intersection of actions, environments, and sense perceptions. We inevitably act within the ecological context of an environment laden with material opportunities for movement and action; in pursuit of these movements, we generate sense perceptions of our surroundings. I also start from the premise that visual perception of the environment arises from active engagement with our surroundings. However, Gibson’s original formulation was organism-agnostic and thus had little to say about how culture or social structure shape the way people engage their environments. A sociological application of affordances requires two conceptual steps, from actions to practices and from environments to places. By taking these steps, we retain affordance theory’s focus on the ecological entanglement of environment, perception, and action while accounting for the social and historical contingencies of human settings and practices. In this regard, affordance theory can equally benefit from engagement with sociological theory.

What affordances afford the sociology of place and of the senses

Affordances heuristically focus attention on the middle-range meeting point of the structural and material, on the one hand, and the phenomenological and pragmatic, on the other. I do not claim to resolve any tensions that exist between these schools of thought, nor am I the first to invoke affordances within sociological treatments of practice and culture. Rather, I hope to underscore three closely related points that together summarize the explanatory and integrative value of affordances for the sociology of place and of the senses.

The first is that many practices—perhaps all practices—are not generated by the social body-habitus in isolation but in dialogue with the specific material elements of one’s environment, including other people. To invoke a famous “carnal” study of the habitus in formation, Wacquant (2006) needed a gym, teachers, and sparring partners to internalize the embodied skills of boxing. Even dispositions seemingly confined to the body, such as posture, only become meaningful in the intersubjective realm of social interaction (Gumperz 1982; White 2008). As such, all require immediate others to make sense qua practice.

Following from this, the second point is that affordances are spatially distributed in ways that entail a coextensive spatial distribution of practices. As the phenomenologist of place Casey (2001, 686–89) notes in his discussion of Bourdieu, “a given habitus is always enacted in a particular place and incorporates the features inherent in previous such places, all of which are linked by a habitudinal bond.” Because boxing is historically a working-class and disproportionately Black sport in the United States—and because prestigious practitioners of ethnography often end up in Chicago—it is unsurprising that Wacquant’s study embedded him within a working-class Black Chicago neighborhood, at a gym frequented predominantly by working-class Black men. In turn, this emplacement yielded both a particular ethnographic vision attentive to the role of the boxing gym in the lives of men coping with racial oppression and urban poverty, and the specific regime of boxing training he underwent, as boxing “styles” are geographically variable and famously racialized (Smith 2023).

The final point is that this ecology of practice entails a parallel ecology of somatic experience. Contra a view of cognition imposing order on fundamentally unordered stimuli, the world of qualia is, at least to some extent, ordered rather than random (Martin 2011, 112–44). As our practices emplace us, they usher us into arrays of ordered phenomena that we attend to via socialized skills of environmental engagement and perception. When environments constituted by sets of affordances encounter people with the socio-somatic skills to engage those affordances in certain ways, a process is set into motion that generates a flux of distinct sensory experiences—the scent of pine trees, the frosty dampness of snow, the weight of acceleration and the whistling of the wind—that collate into meaningful experience, to return to a previous example, of skiing, or even of becoming a skier.

In summary, one way to think about places is in terms of affordances: potentialities of action and perception that inhere in the sociomaterial elements of an environment. Different sorts of places are constituted by different sorts of affordances. Practices are socially stratified and entail skilled, enculturated interactions with affordances. Practices are therefore distributed in space (or, put differently, across places) just as affordances are. While pursuing habitual practices in dialogue with local affordances—while dwelling in place (Ingold 2000)—people generate sensory experiences. Some of these experiences are interpreted or “felt” in terms of apprehended qualities, that is, qualitatively. The ways in which we interpret and apprehend these experiences—as pleasurable, uncomfortable, exciting, empowering, beautiful, alienating, relaxing, and so on—bear upon our sense of self and therefore on our future regimes of practice (e.g., Wohl 2015). These evaluations are not made by isolated minds. They are at least partially contingent upon the habitus and upon historically and socially variable structures of motivation related to (among other things) social status, as are the practices that generate them.

While I am not the first to apply affordances to sociological analysis, much of this work has been limited to the context of isolable technological artifacts in organizational settings (Hutchby 2001; Fayard and Weeks 2014), or to theoretical refinement of the concept for sociology (Davis 2020; see Borghi 2021 for recent parallel work in cognitive science). Jerolmack and Tavory (2014) articulate with great clarity the ways in which constitutive social practices require nonhuman affordances, though the historical structuration of affordances into places—operating both as spatially bounded interpretive frames and material substrata—remains underexplored (though Lewinson [2011] insightfully relates affordances to situated feelings of domestic comfort). Understandably, the affordance concept’s emphasis on enabling movement and action has been more commonly seized upon by social scientists than its emphasis on visual perception. When visuality has been salient, as in McDonnell’s (2010) work on AIDS prevention campaigns in Ghana, it has been the relationship between material properties and appearances of objects (e.g., decaying public health campaign materials) and subsequent actions and meaning-making (e.g., adopting safe-sex practices) that has been explored, rather than the relationship by which actions supported by affordance structures generate structured and meaningful visual experiences. Relating the distribution of UFO sightings to the geographical distribution of affordances, as I do here, is therefore an attempt to leverage and emphasize this visually generative dimension of the latter within the context of generalist sociology.

To conclude, thinking through affordances can help organize sociological accounts of emplacement, practices, and sensory experience. On the one hand, we can ask what historical processes shaped the set of affordances present somewhere. Local affordance structures can be treated as outcomes to be explained, often by reconstructing the actions and decisions of powerful actors with considerable capacity to shape the environment through force and the mobilization of capital (see Molotch et al. 2000 for a rigorous example). On the other hand, we can ask how affordances go on to structure a range of locally possible practices and the sensory experiences their pursuit entails. I visualize affordances as both the crystallization of historical processes and the enablers of practices that generate sensual experience in Figure 2.

Figure 2

Affordances connect sociological thinking about place, embodied practice, and the senses. The intersection of the diagonal lines represents the present moment. Affordance structures—such as a farm (left) or a nature trail (right)—encourage certain practices while constraining others. The source of this encouragement and constraint is material (e.g., swimming requires water) as well as sociocultural. That is, socialization results in bodily skills as well as judgments of (in)felicity linking certain activities to certain kinds of places, i.e., to certain affordance structures but not others.

III. Rural recreational counties

In the previous section, I elaborated a framework, centered on affordances, that focuses on the tight linkages between places, practices, and sensory experiences. Here, I apply this framework to a kind of place that has emerged in the wake of postwar economic globalization: the American rural recreational county. Understanding the rural recreational county in terms of affordances—in terms of its characteristic material possibilities of action and perception—helps to explain the kinds of sensory experiences that disproportionately occur there: among them, UFO sightings.

Until the mid-twentieth century, nearly all rural economies depended on the “extractive” industries of agriculture, mining, timber, and ranching, all of which have since been transformed by automation, globalization, and neoliberal economic reforms. Widespread consolidation of the agriculture industry, for example, led to a 54% decrease in the number of farms from 1910 to 1970, while the average acreage per farm nearly tripled. Ranching consolidated as well—the largest 2% of ranches accounting for 60% of all domestically raised beef—while reforms like NAFTA increased international competition and the offshoring of production. Just as vanishing urban American manufacturing has created gentrifiable urban space in the form of empty warehouses, factories, and lofted workshops, the globalization of extractive industries created vast swaths of rural land available for commercial and residential repurposing (Nelson and Hines 2018, 1479–83).

These changes led to a certain bifurcation among rural places: between those that remained in the extractive paradigm—either as sites of increasing consolidation, or, more likely, of divestment and population decline—and those that pivoted toward what Gosnell and Abrams (2011) call the “post-productivist” industries: recreation, leisure, financial services, real estate, and technology. As a result of these shifts, the 1970s were the first decade in at least 150 years in which the rate of nonmetropolitan population growth exceeded that of metropolitan areas (Johnson and Cromartie 2006). This “rural rebound” continued for the rest of the millennium (Johnson and Beale 1994), not driven by changes to birth or death rates, but predominantly by in-migration from the professional classes of the metropolitan United States (Johnson 2002, 65; Winkler et al. 2007).

What drew this upwardly mobile population to rural areas, who for decades had flocked to cities and suburbs? McGranahan (1999) showed that among several competing hypotheses, the strongest predictor of post-1970 rural population growth at the county level is the presence of “natural amenities”: desirable climate, varied topography, and plentiful water. Counties high on an index of these factors tended to at least double in population from 1970 to 2000, whereas counties low on that index grew an average of only 1% (and half shrunk in population).

As such, the phenomenon came to be known as “amenity migration,” and in 2004 the US Department of Agriculture Economic Research Service added “recreational” to its County Typology Codes to indicate those counties where post-productivist industries had largely replaced traditional rural occupations. Scattered throughout the United States, these roughly 350 nonmetropolitan counties concentrate in the American West, Great Lakes, and Appalachian regions, with smaller pockets in the South and Midwest (Figure 3).

Figure 3

Nonmetropolitan recreational counties as defined by the 2015 USDA County Typology Codes. Metropolitan counties appear in gray. Unshaded areas are nonmetropolitan but not recreational.

To speak of “amenity-rich” areas suggests that natural amenities self-evidently exist. Over the past half-century, however, an interrelated set of embodied practices, moral discourses, and sociomaterial infrastructures concerning the “natural world” have been cultivated to reconstitute the environmental affordances of these places.¹

Materially, this requires capital investment (often from international investors without any connection to the local area; McCarthy and Guthman 1998; McCarthy 2008), part of a more general trend whereby contemporary consumerism shifted its focus from the consumption of goods to the consumption of experiences, lifestyles and leisure (e.g., Pine and Gilmore 1999). The extractive paradigm of economic and ecological relations conceives of environmental affordances as wellsprings of future commodities: cliffsides into ore, forests into lumber, grass and water into beef. Such a paradigm characteristically leads to boom-and-bust patterns of capital investment and divestment that often leaves behind significant environmental degradation (Daum et al. 2019; Mueller 2021). By contrast, the social context of recreation entails a different conception of the environment: now a luxury good advertised to the metropolitan middle classes as an object of aesthetic consumption and recreational engagement. This is often achieved through coordinated planning efforts between local governments, real estate developers, conservationist groups, and financiers.

To take one illustrative example, the towns of Hamilton and Anaconda in Montana’s Bitterroot Valley studied by Bryson and Wyckoff (2010)—Hamilton successful in reorienting toward the recreational economy; Anaconda, unsuccessful—parsimoniously illustrate the redevelopment of rural places as a result (or in anticipation) of amenity migration. Whereas the environmental degradation wrought by heavy mining thwarted Anaconda’s efforts to recast its affordances as aesthetic amenities, local planners and governance in Hamilton successfully attracted capital investment that converted it into a destination town replete with “a brand-name championship golf course; 30 miles of hiking, biking, and horse trails; an equestrian center; and an elk-viewing gazebo that allows club members the opportunity to survey some of the 1,400 acres of conservation easement within the development” (Bryson and Wyckoff 2010, 63).

The reconfiguration of environmental affordances is not limited to outdoor recreational infrastructure. Developers strategically parcel residential land and orient housing to maximize the viewsheds—“the areas of land, water, and other environmental elements that can be seen from a fixed vantage point” (Vukomanovic and Orr 2014, 394)—afforded to housing units. “Hedonic pricing models” estimate the value added to properties as a function of the commodified vistas they afford (Sander and Polasky 2009). This “viewscape fetishism” (Van Auken 2010) often overpowers competing planning efforts aimed at sustainability or non-recreational development, as struggles over land and water management between amenity migrants and other locals demonstrate (Golding 2012; Armstrong and Stedman 2013; Bessette and Mills 2021; see also Krause and Robinson 2017).

Infrastructure encouraging sensory engagement with the environment—whether domestic viewsheds, public trails, or privatized spaces like golf clubs—is paralleled by marketing regimes from real estate and tourism organizations. Tourist bureaus define their mission to “connect with our [visitors] on a sensory level to best engage them and better tell this winter story” (Montana Office of Tourism n.d.); encourage an “embrace [of] the full spectrum of sensations on your springtime getaway … where every sight, sound, smell, taste, and touch is an experience to remember” (Pankow 2024); invite you to “satiate all your senses [f]rom stunning vistas to furry farm animals” (Johnson 2020); provide a “feast for the senses” (Blodgett 2025) described in a “sensory guide” (NPS n.d.) that rates potential sensations on a scale of intensity. These often include prescriptive vignettes of the perfect sensory day, for example, in New Hampshire: “The smell of pine drifts through your open window, a kaleidoscope of colorful leaves greets you around every turn, and the sweet taste of apple lingers as you make your way to your next destination” (Visit New Hampshire n.d.).

Of course, tourism materials are oriented toward outsiders, not residents. We should not presume that the “marketing myths of a symbolic countryside” (Hopkins 1998, 65) reflect how residents (whether seasonal or permanent) understand and act within their local environments. Nevertheless, considerable research has shown that amenity migrants do indeed “approach their properties not as livelihood assets but rather as lifestyle assets” (Abrams and Bliss 2013, 856), prioritizing land uses directed at the aesthetic and recreational potential of natural resources (Kendra and Hull 2005; Matarrita-Cascante et al. 2010; Kondo et al. 2012; Ooi et al. 2015). In their study of amenity migrants, Benson and O’Reilly (2009, 1) summarize motivations for relocation as “an anti-modern, escapist, self-realization project, a search for the intangible ‘good life,’” leading Hines (2010) to liken amenity migration to a state of “permanent tourism” oriented toward the consumption of aestheticized experiences.

These findings do not allow us to generalize to all amenity migrants, and rural recreational areas contain non-migrant populations (e.g., tourists, multigenerational residents, and low-wage hospitality and service laborers). However, regardless of the normative identities of those who find themselves in rural recreational areas, the structure of locally available affordances systematically encourages certain forms of leisurely, sensual engagement with the outdoors, practices that have become assimilated within a broadly upper-middle-class habitus in the last century.

In sum, the ideal-typical recreational rural area is characterized by the presence of several correlated but ultimately distinct features: natural amenities, recreational infrastructure, and an economic base to attract and sustain in-migration from metropolitan areas. From hiking trails and golf courses to kitchen windows and patios, the affordance structures of outdoor and domestic spaces—in tandem with practices culturally associated with rural leisure that depend upon those affordances—facilitate outdoor recreation and a particularly visuo-aesthetic orientation toward the environment. While elites have sought refuge from urban centers in rural estates since the Industrial Revolution (Jacoby 2014) and still do (Farrell 2020), these development strategies have extended identities and practices of leisurely rural luxury to the middle and upper-middle classes.

IV. UFOs in rural America: An affordance-based exploration

In the previous section I described the emergence and nature of the rural recreational area, emphasizing the tight linkages, mediated by affordances, between historical patterns of capital investment and embodied practices of active, sensual engagement now characteristic of those areas. Here, I hope to illustrate, if in exploratory form, what explanatory value can be gained from thinking in this way. To do so, I turn to an empirical puzzle that has long intrigued and frustrated social scientific analysis: the social distribution of firsthand UFO sightings.

Data

Here I turn my attention to the set of mechanisms grounded in localized affordance structures. In the great majority of studies of UFO seers, the actual circumstances and narration of the moment of sighting are entirely absent. While psychological explanators are thoroughly investigated, the body, regular practice, and the ecology of affordances remain invisible. What can a return to the “soil of the sensible” (Merleau-Ponty 1964, 160)—research designs emphasizing the concrete, emplaced, sensate moments from which UFO sightings emerge—offer the social science of UFO experiences, and cultural sociology more generally?

To attempt an answer, I analyze two decades of UFO sightings in nonmetropolitan counties reported to the National UFO Reporting Center (NUFORC) from 2000–2020 (N = 21,062). Established in 1974, NUFORC has documented over 130,000 reports in the United States. In their own words, NUFORC’s “primary function over the past four decades has been to receive, record, and to the greatest degree possible, corroborate and document reports from individuals who have been witness to unusual, possibly UFO-related events.” They claim to be “well known by law enforcement agencies, FAA ARTCCs and flight service stations, National Weather Services offices, military facilities, NASA, and many 911 emergency dispatch centers all across the United States and in many parts of Canada,” who “routinely” direct related inquiries to the center. Indeed, official FAA policy redirects UFO reporters to NUFORC (FAA 2024). Someone wishing to report a sighting does so through a structured webform including several questions with predetermined options as well as text boxes (Figure 4). No personal information is systematically collected.

Figure 4

A typical UFO sighting report in the NUFORC database. Notes in double-parentheses indicate editorial comments.

Specifically, I compare reporting rates in recreational versus non-recreational rural counties. I rely on the USDA’s Rural-Urban Continuum Codes to identify nonmetropolitan counties in the United States. Metropolitan counties either (1) contain an urban center of at least 50,000 people, or (2) are tightly integrated with the former by commuting patterns, regardless of urban population size. All other counties are classed as nonmetropolitan and included (N = 2,057).⁵ I focus only on the nonmetropolitan United States to minimize complications introduced by densely built environments and to reduce overall variation in affordance structures beyond the recreational/non-recreational distinction of interest, though the effect of recreational status is substantively comparable when including metropolitan counties in the models.

Whereas the USDA indicator variable for recreational status is based only on economic profile, I decompose recreational status into the two dimensions that substantively define it: natural amenities and recreational economics. These are correlated but not identical. There are high-amenity counties that have not experienced considerable amenity migration as well as lower-amenity counties with recreational economies. Natural amenities are defined according to the composite variable developed by McGranahan (1999), combining measures of average temperature, sunshine, precipitation, water area and topographical variation. I replicate the methodology used by the USDA (2025) to identify significantly recreational economies—proportional employment in “arts, entertainment, and recreation, accommodation and food services, and real estate and rental and leasing services” as well as the proportion of vacant “seasonal” housing—using data from the 2016 American Community 5-year (ACS5) Census survey. Disentangling these dimensions offers greater analytic granularity than a single binary opposition: a 2 × 2 space isolating both the “natural” and “manufactured” components of contemporary rural recreation, allowing me to assess whether the effects of scenic environments and recreational infrastructure are additive to or redundant with one another. I define counties high in natural amenities and in recreational economics as those in the top quartile of their respective variables; results are substantively unchanged with different thresholds. I refer to counties high in both variables as “ideal-typical recreational.”

To model the expected frequency of UFO reports, I include a suite of control variables including: socioeconomic, racial, and ethnic demographics; housing data; presidential election data; religious membership data; dummy variables indicating proximity to Department of Defense military bases, NASA bases, national parks, and the two most famous sites in American UFO lore (Roswell, New Mexico and Area 51, Nevada); water area; temperature, precipitation, and fog data; light pollution proxied by population density; and air traffic data—specifically, the number of airports within 100 miles of the county centroid, and the yearly number of passengers traveling to and from those airports as a proxy for flight volume. To maximize the demographic data coverage and encourage data reproducibility, Census variables are sourced from the 2016 ACS5 survey. Appendix A fully describes the data construction process, including the decisions that went into defining the observation period, choice of Census years, and robustness checks.

V. Results

Table 1 reports descriptive statistics across the 2 × 2 space of recreational status. I report median figures given the long-tailed distribution of UFO reporting rates, though means yield similar results. These statistics clearly support the general hypothesis: counties low in both amenities and recreational economics are the site of considerably fewer per capita reports (0.24 reports per 1,000 residents) than those in high-amenity counties (0.35) and in high-recreational counties (0.40). Meanwhile, counties high in both indexes are the site of 0.71 reports per thousand residents, a roughly 196% increase relative to the baseline of counties low in both measures. The use of medians suggests that outlier counties are not driving the observed differences.

These measures must be interpreted cautiously. One concern relates to “counting snowbirds” (Happel and Hogan 2002). Recreational counties are characterized by an abundance of what the Census Bureau calls “vacant seasonal” housing: housing owned and periodically occupied throughout the year, as opposed to vacant for sale or rent. About 75% of vacant housing is considered “seasonal” in counties that the USDA classifies as recreational, compared to about 32% in other rural counties. If seasonal residents are elsewhere when the Census is enumerated in April, the “true” population of recreational counties is at risk of underestimation, inflating per capita reporting rates.

I address this counterfactually: what if we assumed that all seasonal homes were occupied year-round? This hypothetical population measure will increase populations in all counties but will overrepresent the effect of seasonal residents (who, by definition, are not always present), providing a conservative estimate of per capita report rates that deflates rates in ideal-typical recreational counties much more so than in others. I estimate this value by multiplying the average number of residents per home by the number of vacant seasonal properties in the county. Normalizing reports by this hypothetical population only slightly shrinks the discrepancy: the median report rate for ideal-typical recreational counties drops from 0.71 to 0.57 but remains 147% greater than the baseline in counties high in neither (0.23), while counties high only in amenities (0.31) or in recreational economics (0.32) show intermediate values. The idiosyncrasies of population enumeration in highly seasonal areas are not meaningfully driving differences in reporting rates.

As an additional robustness test, I ask whether the difference shrinks or disappears when only comparing ideal-typical recreational counties to their non-recreational immediate neighbors. It could be that the geographic distribution of things able to be seen in the sky is skewed in a way unrelated to the present hypothesis (e.g., due to unobserved weather phenomena), but which arbitrarily cluster around recreational areas and their immediate neighbors. If this were the case, reporting discrepancies between ideal-typical counties and their non-recreational neighbors could shrink or disappear.

This is not the case: the 126 counties high in neither natural amenities nor recreational economics that are adjacent to at least one ideal-typical recreational county have a median report rate (0.24 per 1,000 residents) nearly identical to that of the entire set of such counties. In terms of UFO reports, non-recreational rural counties that abut recreational ones are the same as the entire population of non-recreational counties, but different from recreational counties.

Negative binomial regression

The observed variation may derive from demographic differences rather than variation in affordances and associated practices. UFO belief and interest is not limited to any subpopulation, but it is not uniformly distributed, either. Age differences could particularly matter, as younger people are more likely to be able-bodied, expanding their possibilities of environmental engagement compared with elderly populations. Political orientation and religiosity might plausibly matter, too, along with non-demographic variation affecting sky visibility: weather, air traffic, proximity to military bases and airports, and so on. Whether natural amenities and recreational economics remain predictive of UFO reports after controlling for other forms of variation remains to be seen.

To this end, I fit a series of negative binomial models to estimate the number of expected county-level UFO reports. Continuous independent variables are scaled to a mean of 0 and standard deviation of 1. Appendices provide the unabridged regression table for the fully specified model (B), a data dictionary (C), maps visualizing the geographic distribution of residuals (D), predicted versus actual values (E), and a number of alternate specifications: with continuous measurements of natural amenities and recreational economics (F), bivariate models (G), a fully specified model with added state fixed effects (H), and a Bayesian negative binomial model with conditional autoregression to account for spatial relations between observations, with (I) and without (J) state fixed effects. Findings are robust to all alternate specifications.

Table 2 displays the regression coefficients associated with the two measures of recreational status, while Figure 5 displays the same results as the marginal effects of the four possible combinations of the two key variables (i.e., the number of reports predicted by the fully specified model, holding all other variables at their means and varying only the values of the two key variables).

Table 2

Negative binomial regression results, fully specified model.

	Dependent variable: Count of UFO reports, 2000–2020
	(1)	(2)	(3)
High natural amenities	0.343*** (0.040)		0.309*** (0.041)
Recreational economy		0.255*** (0.040)	0.206*** (0.040)
County-level controls	Yes	Yes	Yes
Observations	2,057	2,057	2,057
Log Likelihood	–5,583.890	–5,598.404	–5,570.134
θ	4.086*** (0.217)	3.952*** (0.207)	4.146*** (0.220)
Akaike Inf. Crit.	11,217.780	11,246.810	11,192.270

• Note: *p < .05; **p < .01; ***p < .001.

Figure 5

Marginal effects plot of the two primary independent variables operationalizing rural recreational affordance structures. Counties high in both are the sites of nearly 70% more UFO reports than counties high in neither, when holding all other observed variables at their observed means.

A county high neither in amenities nor recreational economics but average in all other variables is predicted to be the site of about 8.31 reports over the observation period. For counties high only in recreational economic dependence, this predicted value is 10.3, while it is 11.3 for counties high only in natural amenities. Crucially, counties high in both measures but average in all other observed ways are predicted to be the site of about 14.0 reports. The marginal effects of being an ideal-typical recreational county predict being the site of a roughly 68% increase in UFO reports, net of all other observed variation, compared to non-recreational rural counties. Such an effect would be considerably larger than any from the sociopsychological literature on variation in UFO experiences familiar to the author.

Narrative features

In support of the hypothesis, these results suggest that the social context of rural leisure yields much greater rates of UFO sightings than other rural contexts. However, these results could be dismissed by ecological fallacy: just because a county’s natural features and economic base lend themselves to certain practices while constraining others does not mean that individuals see UFOs while pursuing these practices. Statistically, all we can say thus far is that counties with this profile tend to be the site of these reports with greater frequency than we would expect. Perhaps what I interpret as a difference in sighting frequency is actually a difference in likelihood of reporting sightings. If so, differences in socialized modes of environmental engagement could be irrelevant. The situated moments of perception and action through which the UFO encounter unfolds might not meaningfully differ from place to place.

For this reason, reporters’ narratives of UFO sightings are invaluable. Table 3 displays a curated sample of narratives. Narrative data allows us to reconstruct narrators’ experiences and understandings of events: in their own words, what they were doing, why they were doing it, and the subjective states aroused by an event. Reporter narratives are often highly detailed. They get us closer to the soil of the sensible: the situated, practical affairs during which the UFO is seen.

Table 3

Selected report excerpts from recreational and non-recreational counties.

Recreational

Non-recreational

I was looking out my sliding glass door, through the screen, at what appeared to be almost a Christmas card photograph; two trees and a star between them on a slope on top of the ridge. I remarked to my friend that the scene looked like a card, and he looked and saw what I saw … (S62587) My friend and I got into the hot tub to enjoy the cool evening before going in to eat dinner. 10 min later as we were both looking at the western sky I saw what appeared to be a huge flash of light in the sky … (S48383) My husband and I stopped to take photos of Navajo Lake just outside of Duck Creek Village. …We stopped to see the photos and noticed this green sphere in all the photographs. (S115396)

We were in the front pasture because the horses were all lined in a row acting bizarre. We walked out to them, and I looked up at the sky, and there was something not of this world … (S167258) I was getting in truck to leave for work, I had the feeling of static electricty [sic] on my neck. … As i looked around, I looked up, I saw a craft in the air … (S49379) Since I live on a farm with my parents, I have to get up to do chores in the morning, and I do the same at night. During the winter months, it is usually dark at those times, so the night sky can be visible if it isn’t cloudy. On several occasions, I have witnessed a bright yellow-orange light off to both the east and west of our house … (S111609)

I conduct a stratified random sample of 1,000 reports: 500 from ideal-typical recreational counties and 500 from rural counties below the 50th percentile on the amenity scale and recreational economic index. This latter category is a heterogeneous group: not just mining country or America’s breadbasket, but college towns, East Coast hamlets, American Indian reservations, and so on. I anticipate that narrators’ descriptions of what they were doing immediately prior to the sightings will reveal meaningful differences in patterns of sensual attention to and active engagement with different types of environments. I exclude reports shorter than 50 words from the population to reduce the likelihood of sampling uninformative descriptions. Four sampled reports (0.8%) were omitted because they describe secondhand events or were inscrutable.

I predict that two practices—what I call active looking and outdoor leisure/recreation—are likely to occur more frequently in ideal-typical recreational areas and to entail a certain modality of engagement with one’s surroundings that lends itself to UFO sightings. While all UFO encounters involve vision, active looking refers to narratives in which the witness describes themselves as already engaged in an act of visual surveillance prior to the sighting. “We were looking up at the sky remarking how clear it was and how many stars were out …” (report S151849) describes active looking; “Myself and three friends were standing outside having an evening cigarette before retiring for the night when …” (S98913) does not. “I usually look for the moon before closing my curtains. That night …” (S126952) contains active looking, “I was in the process of feeding my dogs when …” (S73372) does not. Of course, that a reporter does not include the language of active looking cannot be interpreted as ironclad proof that they were not visually surveilling their surroundings. Its presence, however, suggests that they understood themselves to be actively doing so.

Outdoor leisure/recreation refers to activities such as fishing, hunting, hiking, bonfires, jogging, going on walks, and soaking in a hot tub. It does not refer to walking one’s dog, smoking, taking out the trash, mowing the lawn, walking to the store, loading a car, or any form of employment. While the absence of these narrative features cannot prove their absence in the actual moment, their presence evidences a particular understanding the narrator brings to their own practices, while their disproportionate presence in a certain type of environment suggests a non-incidental relationship.

In addition to these narrative features, I code reports for several additional features to contextualize any observed trends:

• whether they provide any narrative account of what they were doing at the time of the sighting;

• whether others were present;

• whether they were at home;

• whether they were driving;

• whether they were outside, excluding driving; or

• whether they were working.

All features are coded as 0 if absent from the report and 1 if they are present. Reports without any narrative account—i.e., reports that describe the thing seen, but that provide no description of their whereabouts or activities immediately prior to the sighting—are dropped from calculations of narrative feature prevalence. If significant ambiguity as to the presence or absence of a feature exists, that feature was left uncoded for that observation and the observation was excluded from calculations involving that feature. The only feature for which this was a somewhat regular occurrence was the “at home” feature, with about 5% of ideal-typical reports and 8% of other reports left uncoded. For most other features, less than 1% of reports were uncoded.

Reports from ideal-typical recreational counties and their comparison group are not entirely dissimilar, ruling out certain interpretations of the data. Reporters from each describe what they were doing when the sighting occurred at similar rates (75.5% among ideal-typical counties, 74.6% among others), suggesting that observed differences cannot be attributed to an inherent discrepancy in the amount of detail people provided. Similarly, one could imagine that a thing seen in the sky is more likely to be interpreted as a UFO if someone else is there to see and verify it, associating the likelihood of group activity with the likelihood of reporting. This is not the case: among reports that include narrative context, reporters mention the presence of others at comparable rates (49.7% ideal-typical, 47.6% other). Sightings while the reporter was at work also occur at similar rates (5.6% and 4.9%, respectively).

Two other features exhibit somewhat larger and statistically significant differences. Whereas only 24.6% of reports from ideal-typical recreational counties occur while driving, this number jumps to 37.9% for other counties. Further, 56.3% of recreational reports occur outdoors (excluding driving), versus only 51.9% of non-recreational reports. Both figures are compatible with an affordance-based interpretation. For a sighting to occur, it must be possible to see outside. In non-recreational areas, these conditions are more likely to be met in the context of driving, whereas in recreational areas, other contexts are more likely to present the opportunity to see something in the sky. The fact that driving generally directs one’s sensory attention toward aspects of one’s surroundings relevant to navigation—the road, signage, other cars—may account for some of the observed discrepancies in report rates (i.e., on average people in both types of counties may be able to see the sky the same amount of time, but in non-recreational areas, much more of this time is spent driving, when would-be reporters are visually preoccupied).

The largest percentage point differences, however, exist for the “outdoor leisure/recreation” and “active looking” features (Figure 6). 34.6% of sightings from recreational counties occur when the reporter was engaging in outdoor leisure whereas only 18.9% of sightings from the comparison group occur in such a context. Similarly, 41.0% of events from recreational counties occur during self-described active looking, compared to only 24.8% of reports from the comparison group. One or the other of these narrative features appears in just over half (50.8%) of recreational accounts, a roughly 62% increase from the rate among comparison accounts (31.1%). In sum, a closer look at this sample of narratives supports the interpretation that variation in routines of environmental engagement and perception characteristic of recreational areas, facilitated and encouraged by locally available affordance structures, is related to the social distribution of firsthand UFO experiences.

Figure 6

Prevalence of key narrative features expressing the embodied dispositions and routines associated with the social context of outdoor leisure. Binomial probability confidence intervals shown.

VI. Discussion

These results should not be mistaken for a complete explaining-away of the UFO phenomenon, nor am I suggesting affordance theory as an encompassing metaphysics that can clarify any sociological topic. The distribution of affordances alone does not and cannot fully account for any sociological outcome, as affordances alone do not fully determine social action and perception. It would also be futile to deny the existence of some normatively odd people who claim to have experienced things like UFOs. Finally, it is important to acknowledge the limitations of the data, which do not make any causal claims possible. In particular, the absence of reporter-level data leaves the door open for the possibility of ecological fallacy: inferring conclusions about individual-level outcomes from group-level measures (such as county statistics).

If we think of ecological fallacy not only as a methodological issue in statistics but as encapsulating a set of deeper claims about the relation between social life and place, however, we might argue these claims demand greater scrutiny in the context of the theories explored here. Ecological fallacy is most frequently invoked in the context of group-level data whose groupness is defined by spatial or geographic boundedness. Yet, as I have argued here, people who find themselves in the same place have mutual access to the same or at least overlapping affordances. In other words, the fallacy becomes less fallacious when “group-level” measures are not means of demographic distributions but rather capture something about available affordances that we believe might relate to outcomes of interest, as is the case with the measures of recreational status employed here. The assumption that actors in the same place have roughly equivalent access to that place’s affordances has tacitly underpinned celebrated research designs in social studies of perception, space and place. For example, in their study of the social construction of neighborhood stigma, Sampson and Raudenbush (2004) measure observed disorder (graffiti, broken windows, etc.) against perceptions of disorder among surveyed residents. Their powerful demonstration that racial and economic characteristics better predict perceptions of disorder than observed disorder itself relies on the assumption that residents have effectively equal visual exposure to the physical indices of disorder.⁶

A similar statistical fear is selection bias: that the people who choose to live somewhere may systematically differ from others in ways related both to their choice of residential location and to the outcome of interest. Like ecological fallacy, however, this observation has been treated as a statistical challenge to be controlled away with sophisticated modeling rather than something of substantive interest itself. As Sampson (2011, 239) comments in his critique of selection bias in the context of residence, “there is something odd about the way social scientists approach selection … almost as if they are spooked into thinking that choice [of residential location] renders the environment impotent.” Instead, he suggests, all “individual perceptions and preferences are formed in part as a reaction to the environment,” “all behavior is situated in a context,” and therefore the methodological imperative to control away these endogenous pathways is misaligned with what we know through other means to be true about the complex juncture of people and their social environments (239–40, emphasis original). People who like to ski might be more likely than average to “select into” areas with natural amenities that support skiing. But the inverse is also likely to be true: that the presence of those amenities provides a set of opportunities for action that might make the uninitiated, would-be skier more likely to try their hand at the slopes.

I hope that the research design and observed trends offer proof of concept that thinking with affordances directs our sociological attention toward the entanglement of place, practice, and sensual experience in generative rather than vexing ways: that a return to the concrete, embodied, and sensual circumstances that generate certain social experiences can reveal patterns that are missed by designs targeting abstract beliefs, attitudes, and psychopathology. Because academic social scientists have presumed that psychosocial deviance underlies the distribution of UFO sightings, their research designs have never accounted for the differential ways that people move through and attend to their differential local surroundings, whereas this is exactly the sort of variation that an affordance-based research design emphasizes. Recasting the UFO sighting as an ecological outcome, rather than as an individual dispositional or cognitive one, allows for an account of their social distribution that has eluded purely psychosocial approaches. Against the backdrop of seven decades of psychosocial inquiries into the phenomenon, it is perhaps ironic that places historically produced to attract upwardly mobile consumers—not outcasts or deviants, but occupants of a normatively prestigious social class—appear structurally related to UFO sightings.

To summarize the findings: UFO reports occur in rural recreational counties at far greater rates than other sorts of rural counties, with additive effects from both natural amenities and recreational industry. This remains true even when artificially inflating the seasonal populations of recreational areas, and when examining only the set of non-recreational counties that border high-recreation areas. It remains true when fitting models that control for county-level measurements of sociodemographic and climatological variation previously implicated in UFO sightings. These models reveal both high statistical significance and large marginal effects of ideal-typical recreationality. Finally, reports from recreational areas are much more likely to describe sightings as emergent from the exact modalities of action and perception predicted by an affordance-based approach. This suggests that the findings are not reducible to differences in reporting proclivity but instead appear meaningfully related to the sort of eco-sensual variation that affordances emphasize. Taken together, the hypothesis appears supported. More generally, it suggests that the concept of affordances makes the emplaced generation of somatic experience analytically tractable within quantitative designs, connecting the historical and economic production of place to the moment-to-moment, situated practices that generate sensory experience.

Like many forms of experience, UFO sightings occur through sensual engagement with the environment. This is perhaps so obvious that its general theoretical significance has gone overlooked, leaving the social scientific imagination with theories of experiential generation that are far too deviant-psychologistic and insufficiently aesthetico-pragmatic. If other sociological approaches to uncanny experiences have felt the need to put words like “see” in scare quotes—as though the actual embodied circumstances of perception are trivial—here I have embraced the idea that our experience modulates as we turn our necks. Affordances connect our localized practices to the historical production of places, crooking a neck here, drawing an eye there. In attempting to understand the distribution of experiences, especially stigmatized ones—such as, say, of certain physical symptoms (Au et al. 2022), addictions (Schüll 2012), or criminalized compulsions (Katz 1988)—social science should continue to engage meticulously with the interface linking body and place alongside the more psychological or cognitive approaches to perception that have dominated the sociological imagination. The embodied mechanics that lead every so often to moments of “adventitious fatefulness” (Goffman 1967, 165) are no less social than individual phenomena: they require a constant song-and-dance between people, social skills of perception and action, and local affordances.

VII. Conclusion

In this paper I present a sociological interpretation of affordances that bridges the sociology of place and the sociology of the senses. I have argued that, with some help from social theory, affordances can act as a heuristic to focus research designs, especially quantitative ones, on a chain of relationships: between capital and place, place and sociomaterial infrastructure, infrastructure and the habitual practices it encourages, and, finally, practices and the somatic experiences they produce. This historicized image of situated experience can quickly become intractable; I suggest that affordances can bring it down to the soil of the sensible in ways useful to the practical goals of social research. I hope that others can draw further on this approach to affordances, elaborating research designs that remain proximate to emplaced, sensual experience.

Additional File

Appendices A to J can be found here: https://doi.org/10.16995/tsi.18917.s1

Acknowledgments

I would like to thank Peter Bearman, Adam Reich, James Chu, Gil Eyal, Mario Small, Bruce Western, Isaac Reed, Richard Handler, Adria LaViolette, Geeta Patel, Sam Donahue, Anna Thieser, Ari Galper, Ludo Genovese, Amy Weissenbach, Dian Sheng, Emma Miller, Zheng Fu, Monika Yadav, Charlotte Wang, Seungwon Lee, and Katie Zheng for their contributions and support that considerably improved this paper. Thank you to Dallas & Harris Photography for permission to reproduce the photograph in Figure 1. Thank you to Kelsa Kuchera for the drawing in Figure 2. Any mistakes are my own.

Competing Interests

The author has no competing interests to declare.