An information-based approach to action understanding

The psychophysics of affordance perception: Stevens' power law scaling of perceived maximum forward reachability with an object

The literatures on affordance perception and psychophysics are seminal in the basic study of perception and action. Nevertheless, the application of classic psychophysical methodologies/analysis to the study of affordance perception remains unexplored. In four experiments, we investigated the Stevens' power law scaling of affordance perception. Participants reported maximum forward reaching ability with a series of rods (both seated and standing) for themselves and another person (confederate). Participants also reported a property of the rod set that has been explored in previous psychophysical experiments and changes in equal measure with forward reach-with-ability (length). In all, we found that affordance perception reports (β = .32) were an underaccelerated function of actual changes in reaching ability compared with relatively less accelerated length reports (β = .73). Affordance perception scaled with stimulus magnitude more similarly to brightness perception than length perception. Furthermore, affordance perception reports scaled similarly regardless of the actor (self and other), task context (seated and standing), or idiosyncrasies of the measurement procedure (controlling for distance compression effects), while length perception reports were sensitive to location/distance compression effects. We offer empirical and theoretical considerations, along with pathways for future research.

Does lacking information about your affordances impact your perception of others' affordances? A test of the embodied simulation hypothesis

It has been argued that observers perceive actors' affordances via embodied simulation, that is, first perceiving their own affordance, which serves as a model for the actor's affordance, and then adjusting that model to account for differences between themselves and the actor. If so, then preventing observers from picking up information about their own affordances should cause several effects. Specifically, observers should make more errors about the actor's affordance compared to when the observer is free to pick up information about their own affordance. In addition, judgments about the actor's affordance should align better with the observer's affordance than with the actor's affordance, and increase in error as differences between the observer's and actor's affordances increase. The present study tested those predictions. To do so, observers (participants) made judgments about the farthest distance that an actor (a confederate) could reach. The observer's arms were either free to move or were immobilized by having the participant hold them behind their back. The present results did not support the predictions. The present research introduces a novel means for evaluating the Embodied Simulation Hypothesis, provides initial tests of related predictions, and corroborates prior research. In addition, it motivates important questions about embodied simulation and affordance perception.

Perception of another person's maximum reach-with-jump height from walking kinematics

Humans can perceive affordances (possibilities for action) for themselves and others, including the maximum overhead height reachable by jumping (reach-with-jump height, RWJ). While observers can accurately perceive maximum RWJ for another person without previously seeing the person jump, estimates improve after viewing the person walk, suggesting there is structure in walking kinematics that is informative about the ability to produce vertical force for jumping. We used principal component analysis (PCA) to identify patterns in human walking kinematics that specify another person's maximum RWJ ability, and to determine whether athletes are more sensitive than non-athletes to these patterns. Kinematic data during treadmill walking were collected and submitted to PCA to obtain loading values for the kinematic time series variables on the first principal component. Kinematic data were also used to create point-light (PL) displays, in which the movement kinematics of PL walkers were manipulated using the obtained PCA loading values to determine how changes in body-segment movements impacted perception of maximum RWJ height. While manipulating individual segmental loadings in the PL displays did not substantially affect RWJ estimates, PL displays created by replacing the PCA loadings of a high-jumper with those of a low-jumper, and vice versa, resulted in corresponding reversals of participants' RWJ estimates, suggesting that the global structure of walking kinematics carries information about another's maximum RWJ height. Athletes exhibited greater sensitivity than controls to the kinematic manipulations, indicating that they are better attuned to useful kinematic information as a result of their sport experience.

Perceiving and Remembering Affordances for Others Are Continuous Processes

In information-based approaches, affordances are perceived by detecting information that specifies an animal-environment fit, not by combining perceptions of constituent lower-order properties. Given that detection of such information necessarily occurs over space and time, there is no clear distinction between perception and memory. Rather, perceiving and remembering are continuous processes. Whereas previous research has investigated the continuity of perceived and remembered affordances for the self, we did so with respect to perceived and remembered affordances for others. Participants reported remembered maximum reaching height and remembered anthropometric properties of another person. Remembered maximum reaching height was not reducible to a combination of remembered anthropometric properties. Moreover, remembered maximum reaching height scaled to the reaching ability of the other person and not to that of the perceiver. Both results are consistent with an information-based perspective on perceiving and remembering affordances and demonstrate a continuity between perceiving and remembering affordances for others.

Higher order affordances for reaching: Perception and performance

Affordances are available behaviours emerging from relations between properties of animals and their environment. In any situation, multiple behaviours are available, that is, multiple affordances exist. We asked whether participants could detect means–ends relations among affordances (i.e., higher order affordances) in the context of reaching to a maximum height. We both assessed perceived affordances and evaluated actual reaching ability. In Experiment 1, we co-varied higher order goals (reaching to touch vs reaching to grasp) and the lower order effectors used to achieve the goals (fingertips vs a hand-held tool). In Experiment 2, we varied the lower order posture from which reaching would occur (standing vs kneeling). In both experiments, perceived maximum reaching height reflected relations between lower order means (effectors and postures) and higher order ends (reaching goals), and judgments closely reflected actual performance. We conclude that participants demonstrated prospective sensitivity to higher order affordances for reaching extended across multiple levels of the means–ends hierarchy.

Perceiving nested affordances for another person's actions

Affordances are available behaviors that emerge out of relations between properties of animals and properties of their environment. Affordances are nested within one another. One way to conceptualize this nesting is through a mean-ends hierarchy. Previous research has shown that perceivers are sensitive to hierarchical means-ends relationships when perceiving affordances for their own actions. Affordances are also nested in a social context. We investigated perception of hierarchical mean-ends nesting of affordances for another person's actions. We asked participants to judge the maximum reaching height of another person (the "actor"). Judgments of the actor's maximum reaching height reflected manipulated constraints on the reaching task, suggesting that participants were sensitive (prospectively) to hierarchical relations between lower order affordances and higher order affordances. In addition, the results revealed that judgments scaled to the reaching ability of the actor and not that of the perceiver. We argue that perceivers were sensitive to hierarchical means-ends nesting of affordances for another person across two-levels of this hierarchy, and that perceivers' judgments were based upon perceptual information about the actor's action capabilities, rather than being based upon simulation of perceivers' own abilities.

Perception of Affordance during Short-Term Exposure to Weightlessness in Parabolic Flight

We investigated the role of the visual eye-height (VEH) in the perception of affordance during short-term exposure to weightlessness. Sixteen participants were tested during parabolic flight (0g) and on the ground (1g). Participants looked at a laptop showing a room in which a doorway-like aperture was presented. They were asked to adjust the opening of the virtual doorway until it was perceived to be just wide enough to pass through (i.e., the critical aperture). We manipulated VEH by raising the level of the floor in the visual room by 25 cm. The results showed effects of VEH and of gravity on the perceived critical aperture. When VEH was reduced (i.e., when the floor was raised), the critical aperture diminished, suggesting that widths relative to the body were perceived to be larger. The critical aperture was also lower in 0g, for a given VEH, suggesting that participants perceived apertures to be wider or themselves to be smaller in weightlessness, as compared to normal gravity. However, weightlessness also had an effect on the subjective level of the eyes projected into the visual scene. Thus, setting the critical aperture as a fixed percentage of the subjective visual eye-height remains a viable hypothesis to explain how human observers judge visual scenes in terms of potential for action or “affordances”.

Action possibility judgments of people with varying motor abilities due to spinal cord injury

Predictions about one's own action capabilities as well as the action capabilities of others are thought to be based on a simulation process involving linked perceptual and motor networks. Given the central role of motor experience in the formation of these networks, one's present motor capabilities are thought to be the basis of their perceptual judgments about actions. However, it remains unknown whether the ability to form these action possibility judgments is affected by performance related changes in the motor system. To determine if judgments of action capabilities are affected by long-term changes in one's own motor capabilities, participants with different degrees of upper-limb function due to their level (cervical vs. below cervical) of spinal cord injury (SCI) were tested on a perceptual-motor judgment task. Participants observed apparent motion videos of reciprocal aiming movements with varying levels of difficulty. For each movement, participants determined the shortest movement time (MT) at which they themselves and a young adult could perform the task while maintaining accuracy. Participants also performed the task. Analyses of MTs revealed that perceptual judgments for participant's own movement capabilities were consistent with their actual performance- people with cervical SCI had longer judged and actual MTs than people with below cervical SCI. However, there were no between-group differences in judged MTs for the young adult. Although it is unclear how the judgments were adjusted (altered simulation vs. threshold modification), the data reveal that people with different motor capabilities due to SCI are not completely biased by their present capabilities and can effectively adjust their judgments to estimate the actions of others.

Relating spatial perspective taking to the perception of other's affordances: providing a foundation for predicting the future behavior of others

Understanding what another agent can see relates functionally to the understanding of what they can do. We propose that spatial perspective taking and perceiving other's affordances, while two separate spatial processes, together share the common social function of predicting the behavior of others. Perceiving the action capabilities of others allows for a common understanding of how agents may act together. The ability to take another's perspective focuses an understanding of action goals so that more precise understanding of intentions may result. This review presents an analysis of these complementary abilities, both in terms of the frames of reference and the proposed sensorimotor mechanisms involved. Together, we argue for the importance of reconsidering the role of basic spatial processes to explain more complex behaviors.

Factors that affect action possibility judgments: the assumed abilities of other people

Judging what actions are possible and impossible to complete is a skill that is critical for planning and executing movements in both individual and joint actions contexts. The present experiments explored the ability to adapt action possibility judgments to the assumed characteristics of another person. Participants watched alternating pictures of a person's hand moving at different speeds between targets of different indexes of difficulty (according to Fitts' Law) and judged whether or not it was possible for individuals with different characteristics to maintain movement accuracy at the presented speed. Across four studies, the person in the pictures and the background information about the person were manipulated to determine how and under what conditions participants adapted their judgments. Results revealed that participants adjusted their possibility judgments to the assumed motor capabilities of the individual they were judging. However, these adjustments only occurred when participants were instructed to take the other person into consideration suggesting that the adaption process is a voluntary process. Further, it was observed that the slopes of the regression equations relating movement time and index of difficulty did not differ across conditions. All differences between conditions were in the y-intercept of the regression lines. This pattern of findings suggests that participants formed the action possibility judgments by first simulating their own performance, and then adjusted the "possibility" threshold by adding or subtracting a correction factor to determine what is and is not possible for the other person to perform.

A perceptual-motor deficit predicts social and communicative impairments in individuals with autism spectrum disorders

Individuals with autism spectrum disorders (ASDs) have known impairments in social and motor skills. Identifying putative underlying mechanisms of these impairments could lead to improved understanding of the etiology of core social/communicative deficits in ASDs, and identification of novel intervention targets. The ability to perceptually integrate one's physical capacities with one's environment (affordance perception) may be such a mechanism. This ability has been theorized to be impaired in ASDs, but this question has never been directly tested. Crucially, affordance perception has shown to be amenable to learning; thus, if it is implicated in deficits in ASDs, it may be a valuable unexplored intervention target. The present study compared affordance perception in adolescents and adults with ASDs to typically developing (TD) controls. Two groups of individuals (adolescents and adults) with ASDs and age-matched TD controls completed well-established action capability estimation tasks (reachability, graspability, and aperture passability). Their caregivers completed a measure of their lifetime social/communicative deficits. Compared with controls, individuals with ASDs showed unprecedented gross impairments in relating information about their bodies' action capabilities to visual information specifying the environment. The magnitude of these deficits strongly predicted the magnitude of social/communicative impairments in individuals with ASDs. Thus, social/communicative impairments in ASDs may derive, at least in part, from deficits in basic perceptual-motor processes (e.g. action capability estimation). Such deficits may impair the ability to maintain and calibrate the relationship between oneself and one's social and physical environments, and present fruitful, novel, and unexplored target for intervention.

Egocentric or allocentric frameworks for the evaluation of other people's reachability

The adoption of egocentric and allocentric frameworks in the perception of other people's reachability was investigated. In study 1, 24 adults (12 experienced and 12 inexperienced dealing with children) judged vertical reachability for themselves and for two children. In study 2, 37 parents judged vertical reachability for themselves and their children. Absolute errors (|estimate-actual reachability|), absolute percent errors (|1-judgement/actual reachability|×100), and error tendency (underestimations, right judgments, or overestimations) were calculated. Adults were quite accurate in perceiving their own reachability (absolute percent errors ranging from 2.20% in study 1 to 3.12% in study 2) and clearly less precise when estimating children's reachability. Results indicated a tendency for adults to overestimate reachability of the younger child (study 1) and a tendency for parents to overestimate their children's reachability (study 2). No correlation between judgement errors for the self and for the children in any of the studies was observed. Results support the existence of an allocentric and not an egocentric framework when evaluating other people's affordances.

Perceiving affordances for joint actions

Two individuals acting together to achieve a shared goal often have an emergent set of afforded behavioral possibilities that may not easily reduce to either acting alone. In a series of experiments we examined the critical boundaries for transitions in behavior for individuals walking through an aperture alone or alongside another actor as a dyad. Results from experiment 1 indicated that an intrinsically scaled critical boundary for behavioral transitions was different in individuals than in dyads performing a similar task. Experiment 2 demonstrated that observers are perceptually sensitive to the difference in action parameters for the dyad, while still maintaining perceptual sensitivity about the boundaries of action relative to individuals. In experiment 3, we determined that observers' perception of critical action boundaries for individuals and dyads has a similar informational basis (eye-height scaling). In experiment 4, we demonstrated that observers were able to perceive critical action boundaries for other dyads independently of membership. Together, these results suggest that individuals are sensitive to the affordances related to a joint action, and that this process may not entirely reduce to the perception of the affordances for each individual.

Whom to ask for help? Children's developing understanding of other people's action capabilities

We often rely on other people’s help to accomplish tasks and to attain goals. People, however, differ in their physical action capabilities. Some persons are therefore better able to provide help than others. We investigated 2.5-, 3.5-, and 5-year-old children’s ability to take other person’s action capabilities in a helping situation into account. To this end, they observed a protagonist who needed the help of friends to accomplish several tasks. For each task, two friends were available, but only one was physically able to provide the help. Children were asked to indicate, which partner the protagonist will ask for help. Our results showed a developmental effect with children in the older two groups performing significantly better than those in the youngest group. Additionally, we found evidence that the 5-year-olds outperformed the younger age groups in their ability to justify their choice. Our findings thus suggest that children’s ability to consider others’ physical action capabilities in helping situations develops around 3 years of age. The results are interpreted in terms of children’s ability to perceive others’ affordances. The implication of these findings for theories on the development of action understanding and joint action are discussed.

The influence of athletic experience and kinematic information on skill-relevant affordance perception

Humans can perceive affordances both for themselves and for others, and affordance perception is a function of perceptual-motor experience involved in playing a sport. Two experiments investigated the enhanced affordance perception of athletes. In Experiment 1, basketball players and nonbasketball players provided perceptual reports for sports-relevant (maximum standing-reach and reach-with-jump heights) and non-sports-relevant (maximum sitting height) affordances for self and other. Basketball players were more accurate at perceiving maximum reach-with-jump for another person than were nonbasketball players, but were no better at perceiving maximum reach or sitting heights. Experiment 2 investigated the informational basis for this enhanced perceptual ability of basketball players by evaluating whether kinematics inform perceivers about action-scaled (e.g., force-production dependent), but not body-scaled (i.e., geometrically determined), affordances for others, and whether basketball experience enhances sensitivity to kinematic information. Only basketball players improved at perceiving an action-scaled affordance (maximum reach-with-jump), but not body-scaled affordances (maximum standing-reach and sit) with exposure to kinematic information, suggesting that action-scaled affordances may be specified by kinematic information to which athletes are already attuned by virtue of their sport experience.

Adults' perception of children's height and reaching capability

This study investigated the influence of some characteristics of the task, the model, and the observer, in the estimation errors of adults while judging children's affordances. One hundred and eighteen adults, divided in 4 height groups, estimated height and vertical reaching capability of 3 girls (3.55-, 4.74- and 7.06-years old), in the presence and in the absence of the model. Constant errors (CE) (estimation-real value), absolute percent errors (APE) (/1--estimation/real value/ x 100), and error tendency (underestimations, right judgments, or overestimations) were calculated. A model and a condition effect were verified on APE. APE for the younger model were greater than for the other models (p<0.001), and APE in the absence of the model were greater than in her presence (p<0.05). Generally, adults underestimated height (51.8% of underestimations vs. 32.3% of overestimations) and overestimated reachability (51.3% of overestimations vs. 37.7% of underestimations). The overestimation of reachability was more notorious for the younger model, which might reflect adults' difficulty to consider the specificity of younger children's body proportions. Actually, the overestimation bias may suggest that adults perceive young children as on the basis of adult's geometrical proportions.

Carrying the height of the world on your ankles: encumbering observers reduces estimates of how high an actor can jump

The authors investigated how changes in action capabilities affect estimation of affordances for another actor. Observers estimated maximum jumping-reach height for themselves and another actor. Half of the observers wore ankle weights that reduced their jumping ability. The ankle weights reduced estimates of maximum jumping-reach height that observers made for themselves and for the other actor, but only after observers had the opportunity to walk while wearing the weights. Changes in estimates closely matched changes in actual jumping-reach ability. Results confirm and extend recent investigations that indicate that perception of the spatial layout of surfaces in the environment is scaled to an observer's capacity to act, and they link that approach to another embodied cognition perspective that posits a link between one's own action capabilities and perception of the actions of other agents.