When humans estimate the physical size of another person, they do not need to observe the entire body to make an accurate guess. A new study reveals that people rely heavily on specific combinations of features located in the lower body, particularly around the thighs. The research was recently published in BMC Biology.
Understanding how people perceive bodily dimensions has practical implications for health and psychology. Inaccurate judgments of one’s own shape or the shape of others play a central role in body image disturbances. Conditions like anorexia nervosa are often accompanied by a distorted visual perception of bodily proportions.
Researchers want to understand the exact visual mechanics that lead to these daily misperceptions. To do this, they borrow techniques from vision science to evaluate how visual systems prioritize certain physical traits. Lia Marinko and her colleagues at the University of Western Australia investigated which exact anatomical regions the human brain needs to see to judge weight.
Scientists have debated whether the brain processes a human body as a single object or as a collection of separate parts. The single-object approach is known as holistic processing. The parts-based processing approach suggests we use specific local visual cues to form an overall judgment of the entire person.
Previous studies into this topic have yielded conflicting answers. Some experiments indicate that the overall silhouette is strictly necessary for the human brain to function accurately. Others point out that people tend to fixate entirely on the lower torso when guessing how much someone weighs.
Marinko and her colleagues designed visual tasks to measure two common types of inherent perceptual errors. The first error is known as regression to the mean. This happens when a person judges an extreme object and mentally pulls it toward the average of all similar objects they have seen before in their lifetimes.
For example, people routinely underestimate the weight of very large bodies and overestimate the weight of very small bodies. The second error is called serial dependence. This is a visual illusion where the appearance of an object is distorted toward the object viewed just moments prior.
If a person looks at a sequence of bodies, their judgment of the current body is shifted depending on whether the previous body was larger or smaller. Measuring these two types of innate visual biases allows researchers to track how specific body parts affect the overall accuracy of size estimates.
In the first phase of the study, the research team recruited 99 female participants to look at standard images of female bodies on a computer screen. The participants were mostly healthy young adults. They viewed 35 different bodily forms ranging from extremely thin to extremely large.
To avoid distractions, the faces were removed from the images. During a visual test called the bodyline task, participants saw a body image flash on the screen for a quarter of a second. They then clicked on a flat visual scale from one to seven to rate the perceived size of the model.
The task included extreme body anchors on either side of the visual scale to present a reference point. The researchers divided the computer trials into three different viewing conditions to manipulate the visual information the participants received. The order of the tasks was counterbalanced across participants.
First, participants judged the whole body over a series of trials. Next, they judged images where only the top half of the body was visible, spanning exactly from the navel up to the neck. Finally, they judged images showing only the bottom half of the body, from the navel down to the feet.
The results revealed that estimating size based only on the top half of the body produced substantial perceptual errors. Participants made massive regression toward the mean misjudgments when the lower body was hidden. They were far less capable of telling small and large bodies apart using torso and arm cues alone.
However, viewing the bottom half of the body led to estimates that were just as accurate as viewing the entire body. The estimation errors for the bottom-half images completely mirrored the baseline errors from the full-body images. This result suggests that the human visual system uses a parts-based processing method for body size.
People extract the necessary information completely from the lower extremities. Seeing the upper torso, arms, and shoulders adds virtually nothing to the accuracy of a size estimate based on regression calculations. The brain simply relies on the hips and legs to deduce the rest of the physical shape.
Knowing that the lower body holds the key to visual estimation, the researchers conducted a second experiment to isolate specific leg features. They recruited a new group of 116 female undergraduate students to perform a nearly identical task. This time, the viewing conditions restricted the image down to highly specific sections of the thighs.
Participants viewed the whole body in one condition as a control baseline. In another condition, the researchers applied digital rectangular masks to the images, exposing the outer thigh and hips while obscuring the inner legs. In the third condition, a different digital mask exposed the inner thigh area without showing the outer hips.
The goal was to test if a single isolated cue drove the high accuracy found in the first experiment. The researchers wanted to know if the brain relied solely on the curved outline of the outer hips or the visible gap between the inner thighs. Vision tracking research has previously suggested both of these distinct anatomical areas attract the eye during sizing tasks.
The results of this second test showed that looking at either the isolated inner thigh or the isolated outer thigh resulted in heavy estimation errors. The isolated sections produced staggeringly high error rates compared to the whole body imagery. The single aspects failed to provide enough structural information on their own.
Both the regression to the mean bias and the serial dependence bias inflated immensely under these restricted views. Errors occurred regardless of whether the body depicted on the monitor was very small or very large in physical reality. The researchers noted slightly higher error rates when participants viewed only the inner thigh compared to the outer thigh.
This outcome tells scientists that the brain requires multiple features from the lower body simultaneously to make competent judgments. A person needs to see the outer hip width in relation to the inner thigh contours to make an accurate estimation. A single isolated trait is essentially useless without its neighboring anatomical context.
One complication the researchers noted involves how actual body anatomy changes across sizes in real environments. Smaller bodies tend to feature a visible gap between the upper legs, while larger bodies often have thighs that consistently touch. Despite this physical difference, obscuring either of the thighs ruined the size estimation accuracy for all body sizes equally.
The researchers mentioned a few caveats regarding their specific digital methodology. Applying rectangular masks to obscure body parts is a difficult scientific tactic because erasing sections of an image removes natural cues ranging from skin texture to shadows. The loss of these subtle contextual visual hints might contribute to the elevated error rates.
The study also focused exclusively on female participants looking at female bodies to create a homogenous dataset. Society harbors vastly differing ideals surrounding physical shape based on gender expectations. Men often idealize a muscular shape characterized by a broad upper body and defined chest muscles rather than thigh configurations.
Because of these differing societal standards, people likely view and judge male bodies using an entirely different set of visual rules. Replicating the study setup with male body images and male participants might reveal that upper body features dictate accuracy in that specific demographic. The findings cannot be mapped directly onto differing body types with absolute certainty.
Future investigations could apply these specific visual tests to clinical psychology populations. Exploring how people formally diagnosed with anorexia or binge eating disorders perform on thigh-based visual tests could yield helpful psychiatric insights. A failure to mentally integrate multiple lower body features could explain the severe body distortions experienced by those specific patients.
The study, “The thighs have it: evidence for the importance of lower body regions in female body size judgments,” was authored by Lia Marinko, Briana L. Kennedy, Kei-Kei Koh, Laura Dondzilo, and Jason Bell.
