Tuning in to Another Agent’s Action Capabilities
Tehran J. Davis ([email protected])
Department of Psychology, ML 0376, 429 Dyer Hall
Cincinnati, OH 45221 USA
Verónica C. Ramenzoni ([email protected])
Department of Psychology, ML 0376, 429 Dyer Hall
Cincinnati, OH 45221 USA
Kevin Shockley ([email protected])
Department of Psychology, ML 0376, 429 Dyer Hall
Cincinnati, OH 45221 USA
Michael A. Riley ([email protected])
Department of Psychology, ML 0376, 429 Dyer Hall
Cincinnati, OH 45221 USA
Abstract
Successful social coordination usually requires one person to
be able to judge the limits and abilities of another person.
Two experiments investigating perception of the maximum
height to which an actor could jump and reach were
conducted. In both experiments those estimates were
compared to estimates of the perceiver’s own maximum
jump-with-reach height. Results of Experiment 1 showed that
perceptions for the self improved over time, in the absence of
explicit feedback, but perceptions for the other actor remained
constant. Experiment 2 demonstrated that providing
perceivers with perceptual information about the other actor’s
action capabilities enabled perceivers to recalibrate their
perceptual reports for the other actor’s maximum jump-withreach height.
Keywords: Action understanding; embodiment; affordances;
visual perception; social perception-action.
Most naturally occurring human behavior takes place within
a social environment. In these settings, choosing an
appropriate action for oneself may depend upon the actions
of others. Whether cooperating to lift a couch, or competing
in a tennis match, successful social coordination often
demands that one be able to predict another's actions and
their consequences. One perspective on how agents solve
this problem—a perspective motivated by the discovery of
the mirror neuron system—emphasizes the role of embodied
simulation processes that enable observers to interpret
others’ actions via their own neural systems for movement
(e.g., Gallese, 2005; see also Sebanz, Bekkering, &
Knoblich 2006).
A second (but perhaps not mutually exclusive)
perspective is that action prediction and social coordination
may call upon one's ability to perceive affordances
(opportunities for action; Gibson, 1986) for another
person—one must be able to apprehend what another person
is capable of doing before being able to predict or
understand another person’s actions (Ramenzoni, Riley,
Shockley, & Davis, 2008). In the present research we
addressed two specific questions related primarily to this
second perspective: (1) How does perception of affordances
for another person change over time and with visual
experience, and (2) How does the ability to perceive
affordances for another person depend upon one’s own
action capabilities, as predicted by embodied simulation?
A number of studies have established that people are able
to make accurate perceptual estimates of what the
environment affords others, as well as what the environment
affords themselves, by making use of information that is
publicly available in the optic array to all observers in a
given setting (Mark, 2007; Ramenzoni, Riley, Davis, &
Snyder, 2005; Ramenzoni, Riley, Davis, Shockley, &
Armstrong, in press; Stoffregen, Gorday, Sheng, & Flynn,
1999).
Much of this research has focused on what are called
body-scaled affordances. Body-scaled affordances, such as
sitting and reaching, can be described by referring only to
geometric attributes of the actor (leg length, arm span, etc.)
(e.g., Mark, 1987; Warren, 1984). This class of affordances
represents only a small spectrum of the possibilities for
action that may be available to an individual at any given
moment, however. Recently, there has been increased
interest in so-called action-scaled affordances that are
defined by dynamic properties of the actor (i.e., force
generation, acceleration, etc.; see, e.g., Cesari, Formenti, &
Olivato, 2003; Fajen, 2007; Konczak, Meeuwsen, & Cress,
1992; Oudejans, Michaels, Bakker, & Dolné, 1996; Pepping
& Li, 1997, 2000). For instance, the ability to catch a fly
ball depends upon, among other things, the acceleration and
velocity of the approaching ball, as well as the individual’s
ability to accelerate to the point where the ball will land
(Oudejans et al.).
Action-scaled affordances present a particularly appealing
case for the study of the social perception-action. The
perception of action-scaled affordances for others is
contingent upon the perceiver’s sensitivity to properties of
other person’s action system that may not be obviously
2462
available in the optic array (e.g., the ability to exert force to
lift an object), in the sense that the actor properties such as
these are not simply geometric ones that can be visibly
compared to the complementary geometric property of the
environment.
Previous work exploring the ability to perceive actionscaled affordances for another person (i.e., jump-withreaching) has shown that observers are able to do so with a
similar level of accuracy as perceiving the same affordance
for themselves (Ramenzoni et al., 2005). Ramenzoni, Riley,
Davis et al. (in press) found that perceivers are sensitive to
changes in an actor’s maximum jumping height (effected by
attaching ankle weights to the actor) when the only visual
information about this change is available in the kinematic
patterns of the actor’s gait. In another study Ramenzoni,
Riley, Shockley, and Davis (2008) found that perception of
affordances for another person was influenced strongly by
the perceiver’s own action capabilities—attaching ankle
weights to the perceiver produced reductions in estimates of
an unencumbered actor’s jumping height. Taken together
these findings seem to indicate that the ability to perceive
affordances for others is shaped by sensitivity to the other
person’s as well as one’s own action capabilities.
A number of questions remain unanswered about the
relative influences of the observer’s own action capabilities
on perception of what another actor is capable of doing. For
example, one would expect that perceptual experience or
exposure to new information might produce a change in
affordance judgments through processes of perceptual
attunement or calibration (Fajen, 2005). This raises the
question of whether perceptual reports for one’s own and
another’s maximum jumping height show a similar pattern
of change over time or when new information about the
other person’s action capabilities is provided to the
perceiver. The present study sought to address these issues.
In Experiment 1 we explored whether estimates of the
maximum height one could reach by jumping improved in
accuracy over time as a result of calibration to information
that specifies the affordance. In the absence of new
information about the other person’s action capabilities, a
parallel pattern of change for self and other-actor affordance
judgments would suggest that perception of another
person’s actions is dependent on one’s own action
capabilities. In Experiment 2 we determined whether
watching an actor perform an action that offered perceivers
the opportunity to attune to specifying information for the
actor’s jumping ability could lead to an improvement in
estimating the actor’s maximum jump-with-reach height.
produced by postural sway is sufficient for recalibrating an
observer’s estimations of affordances for the self when the
observer’s action capabilities have been modified. The
present experiment was conducted to determine whether this
finding transfers to perceptual judgments for another actor.
The experiment also permitted an examination of the
relative dependence or independence of affordance
estimates for another actor on the observer’s own action
capabilities. If, as in the Mark et al. study, the observer’s
estimates for the self improved in accuracy over time but
estimates for the other actor did not, it would suggest some
degree of independence.
Method
Twenty participants ranging in maximum jump-with-reach
height from 209.5 to 294 cm (mean = 248.1 ± 24.95 cm)
reported to the laboratory in pairs. On a given trial, each
member of the pair was assigned the role of either observer
or model. Observers were asked to make estimates of the
limits of their own vertical jump-with-reach or that of the
actor when jumping to grasp a small (5 cm × 4 cm)
cylindrical object suspended from the ceiling (see Figure 1).
Jump-with-reach was defined as the maximum height at
which the cylinder could be reached while performing a
vertical jump (i.e., no steps were allowed) with the preferred
arm extended overhead. This definition was provided to
participants prior to beginning the experiment, but
participants did not perform the actual jumping task until all
perceptual judgment trials had been completed. Participants
therefore did not see the other member of the pair perform
the task (reach with jump) until the experiment was over,
but they did have opportunities to observe each other
walking around the laboratory during the experiment before
data collection occurred (cf. Ramenzoni, Riley, Davis et al.,
in press).
Experiment 1
In Experiment 1, we examined whether estimates of
maximum jump-with-reach for the self and the other actor
would improve in accuracy over time, in the absence of
explicit feedback about the accuracy of the estimates. Mark,
Balliet, Craver, Douglas, and Fox (1990) have shown that
even in the absence of explicit feedback about affordance
judgments visual information obtained from the optic flow
Figure 1: Apparatus used in both experiments and depiction
of the perceptual task of estimating the maximum jumpwith-reach height for the self and the other actor.
2463
In all conditions observers stood 3 m away from a vertical
surface in front of which was suspended the target cylinder.
The model was stationed directly beside the apparatus (from
the observer’s perspective, to the right side) at all times.
Both the observer and the model were instructed to remain
stationary with the arms at the sides for the duration of the
experiment. The model was blindfolded and wore
headphones playing music to prevent being influenced by
hearing the perceptual estimates given by the observer.
Perceptual reports were obtained using the method of
adjustments (Guilford, 1954). Participants verbally
instructed an experimenter who, while standing out of sight
behind the apparatus, alternately raised or lowered the
object by means of a rope and pulley until the observer
indicated the object was just at the perceived jump-withreach height for the self or for the model, depending on the
condition. Observers were allowed to fine-tune their
responses until they were satisfied. Between estimates,
observers closed their eyes while the experimenter reset the
apparatus. Successive ascending-descending estimates were
averaged as one trial. Each participant completed six such
trials for self-estimates and six trials for other-model
estimates, with self-and other trials presented in alternating
blocks. Block order was counterbalanced across
participants.
When all trials were completed for the first participant,
members of the pair switched roles. At the conclusion of the
experimental trials, each participant performed three jumpwith-reach actions in order to empirically determine each
participant’s action capabilities.
Results and Discussion
Overall, participants were fairly accurate but exhibited a
tendency to underestimate jump-with-reach ability for the
self (mean estimate = 237.00 cm; mean error = mean
estimate - actual jump-with-reach height = -11.12 cm) and
for the other actor (mean estimate = 234.11 cm; mean error
= -14.02 cm). Ratios of the raw estimates to the participants’
empirically determined actual jump-with-reach heights were
calculated with self-estimates scaled by the observers’ own
action capabilities and other-estimates scaled by the model’s
capabilities. The ratios were near unity for both self- and
other-estimates (self: 0.96; other: 0.95).
To determine the stability of these estimates over time, an
analysis of variance (ANOVA) was conducted on the
estimates with judgment type (self vs. model) and trial
(trials 1 through 6) as within-subjects factors. The analysis
revealed a significant main effect for trial, F(5,95) = 4.62, p
< .05, ηp2 = 0.20. The judgment type × trial interaction was
also significant, F(5,19) = 4.06, p < .05, ηp2 = 0.17 (see
Figure 2). Simple-effects tests revealed a significant effect
of trial for self-estimates, F(5,95) = 7.94, p < .05, ηp2 =
0.18, but not for other-estimates. Estimates increased and
became more accurate over time for the self, but estimates
for the other actor remained constant over time.
Figure 2. Results of Experiment 1. Mean perceptual
estimates of maximum jump-with-reach height for the self
and for the other participant as a function of trials. Bar at
right represents the mean actual maximum jump-with-reach
height across all participants. Error bars correspond to one
standard error.
The difference in the self- and other-estimates over time
suggests that the ability to perceive possibilities for action
afforded by the environment for oneself might be
independent of the ability to perceive behaviors afforded for
others. Had the percepts been strongly dependent, they
would have exhibited similar patterns of change over time.
The relative independence of the percepts could be
explained by differences in the nature of the information
that specifies what is afforded for the self versus what is
afforded for the other. Differences in the relative availability
of such information might be responsible for the observed
results; that is, the lack of improvement in the estimates for
others may reflect a lack of information specific to the other
agent's ability to jump-with-reach. After all, the other actor
simply stood beside the apparatus, without jumping or even
extending the arms overhead. While body sway might have
provided participants with enough information about their
own relation to the environment to allow them to tune to the
information that specified their own maximum jump-withreach (cf. Mark et al., 1990), a shift to a different source of
information about the other model’s relation to the
environment might be necessary to allow for an
improvement in estimating maximum jump-with-reach for
the other person.
In Experiment 2 we explored this possibility. We
investigated whether participants’ estimates of maximum
jump-with-reach height for the other person changed if
access to additional information about the action capabilities
of the other person was provided to the observer. Rather
than simply allowing the observers to watch the other
person jump, however, we allowed observers to watch the
model perform a different task that was functionally related
(but not identical) to jumping—squatting to lift a weight.
2464
Experiment 2
Previous findings have demonstrated that information
contained in kinematics of a model’s movements is
sufficient to inform observers about the model’s action
capabilities (Ramenzoni, Riley, Davis et al., in press).
According to the kinematic specification of dynamics
(KSD) principle (Runeson, 1983; Runeson & Frykholm,
1983; Runeson, Juslin, & Olsson, 2000) the kinematics and
dynamics of an event are specific to one another (i.e., they
relate in a 1:1 fashion because the forces and masses
lawfully determine the resulting motion according to
Newton’s laws), from which it follows that kinematic
information is a reliable source of information for
perceiving facts about the underlying dynamics of an event.
The kinematics of a person’s movement patterns—the
patterns of displacement, velocity, and acceleration of the
limbs and body segments—thus may be informative about
the person’s action capabilities—the person’s dynamic
capacity to produce force for jumping. How high a person
can jump is determined by an actor’s ability to produce
vertical force impulses relative to the actor’s mass. Based on
these considerations, we hypothesized that providing
perceivers with information about a person’s dynamic
capabilities—even in the context of a behavior that differed
from the behavior about which perceivers made judgments,
namely a lifting (squatting) task—would enable perceivers
to tune in to the model’s action capabilities and improve
their estimates of what is afforded the other actor.
Perceivers initially provided two sets of estimates before
watching the model perform the lifting task. Then they
watched the model repeatedly lift (squat) 10% of his body
mass. We expected that observing the actor perform the
lifting task could provide information about the actor’s
ability to produce force with the legs in order to raise a mass
(Dowling & Vamos, 1993), and therefore provide
information about the actor’s ability to lift his own body
from the ground during a jump. Estimates provided for the
model before and after watching the model lift the mass
were compared to estimates provided by a second group of
control participants who observed the actor perform a task
that did not share underlying dynamics with jumping—
standing while rotating the torso (twisting about the waist).
Based on the results of Experiment 1, we expected to
observe a significant improvement in perceptual estimates
over time for self-estimates for both the experimental and
control groups. We furthermore expected the other-model
estimates to become more accurate for the experimental
group only, and not simply with the passage of time but
only after watching the model perform the lifting task.
Method
Thirty participants, ranging in height from 151.5 cm to 196
cm (mean = 170.6 ± 11.12 cm) and in maximum jump-withreach height from 220.2 cm to 298 cm (mean = 252.86 ±
21.58 cm), were asked to make jump-with-reach estimates
for themselves and a model. The model was 180 cm tall and
weighed 77.27 kg, with a maximum standing reach of 215.1
cm and a maximum jump-with-reach height of 280.2 cm.
The apparatus from Experiment 1 was used along with a
134 cm long, 11 cm diameter tube containing weights
equaling 10% of the model’s body mass (8 kg), used by the
model when performing the squatting task.
The procedures for the second experiment closely
followed those in Experiment 1. Participants made estimates
about the limits of their own vertical jump-with-reach, as
well as for the model. Estimates were again obtained using
the method of limits. Participants were given a block of 8
practice trials for each judgment type to familiarize them
with the task. Trials were presented in blocks of 4
successive estimates for the self and 4 successive estimates
for the model. The order of judgment type conditions (self
vs. model) was counterbalanced across blocks and across
participants. Two blocks of each estimate condition were
administered, for a total of 16 trials.
After the first block of experimental trials both the
participant and model were led into an adjoining room. The
participant and model faced one another, standing 2 m apart.
The participant was instructed to watch the model perform
either the squatting or torso rotation task, depending on the
group assignment for the participant. In the control
condition, the participant observed the model who stood
rotating his torso from one side to another a total of 15
times. The rotation task did not involve any flexionextension about the ankle, knee, or hip—effectively the
model stood upright, with the legs straight, and twisted
about the waist while holding the tube at chest level with
both hands. In the experimental condition, the participant
observed the model lifting the tube using a squatting
technique. In this condition, the model held the tube at chest
level with both hands. While holding the tube, the model
lowered his torso by bending at the knees and keeping his
back straight until his knees were flexed 90˚ and his upper
legs were parallel to the floor; at this point the actor stopped
and raised himself back to a standing position. The model
lifted the tube in this manner a total of fifteen times. Just
prior to observing the model, every participant lifted the
tube once. At the completion of the model’s task, both the
participant and model returned to the other room and
completed the remaining block of estimates.
Results and Discussion
Estimates from the four repeated trials in each condition for
each subject were averaged to yield one mean estimate per
condition per subject for both the experimental and control
groups (see Figure 3). For the control group there was a
tendency to underestimate maximum jump-with-reach
height for the self (mean estimate = 241.40 cm; mean error
= mean estimate - actual jump-with-reach height = -12.99
cm) and the model (mean estimate = 256.02 cm; mean error
= -24.18 cm). A similar tendency was found for the
experimental group in the estimates made for the self (mean
estimate = 238.41 cm; mean error = -18.06 cm) and the
model (mean estimate = 250.08 cm; mean error = -30.10
2465
cm). The control group showed greater accuracy for both
self and model estimates compared to the experimental
group. Ratios of the raw estimates to the participants’ actual
jump-with-reach heights were calculated with self-estimates
scaled by the observers’ own action capabilities and otherestimates scaled by the actor’s capabilities. The ratios were
at or near unity for both groups. As suggested by the error
data, the ratios for the control group (self: 0.95; other: 0.91)
indicated they showed overall slightly higher accuracy than
the experimental group (self: 0.93; other: 0.89).
Figure 3. Results of Experiment 2. Mean perceptual
estimates of maximum jump-with-reach height for the self
and for the other model as a function of block for the control
and experimental groups. Bars at right represent the mean
actual maximum jump-with-reach height for the participants
(self) and the model (other). Error bars correspond to one
standard error.
ANOVA was performed with judgment (self vs. other)
and block (first block—before watching the model—and
second block—after watching the model) as within-subject
factors, and group (experimental vs. control) as a betweensubjects factor. The analysis revealed significant main
effects for judgment, F(1,28) = 21.93, p < .05, ηp2 = 0.44,
and for block, F(1,28) = 27.92, p < .05, ηp2 = 0.50. The
group × block [F(1,28) = 6.80, p < .05, ηp2 = 0.20] and the
judgment × block × group [F(1,28) = 5.19, p < .05, ηp2 =
0.16] interactions were also significant.
Follow-up ANOVAs were performed to further
investigate the significant three-way interaction. ANOVA
was performed on the data separately for each judgment
type. For the self-estimates there were significantly larger
(i.e., more accurate) estimates for the second block
compared to the first block, F(1,28) = 11.28, p < .05, ηp2 =
0.29, regardless of group. Analysis of the estimates
provided for the model also revealed a main effect for block
[F(1,28) = 15.16, p < .05, ηp2 = 0.35] and also a significant
block × group interaction, F(1,28) = 6.31, p < .05, ηp2 =
0.18. The interaction was driven by experimental group
participants providing significantly larger estimates for the
model in the second block compared to the first block, t(14)
= -5.95, p < .05. No differences between blocks were found
for the estimates provided by the control group.
The results showed that participants’ estimates of
maximum jump-with-reach height for the self became more
accurate in the second block compared to the first block for
both the control and experimental groups. Such an increase
in accuracy was predicted based on the results of
Experiment 1. The control group’s estimates for the model’s
maximum jump-with-reach height did not change across
blocks, also similar to Experiment 1. However, the
experimental group participants’ estimates were more
accurate in the second block than in the first. Watching the
model perform a task related to the dynamics of jumping
(experimental group) helped perceivers tune in to the
model’s action capabilities. Watching the model perform an
unrelated action (control group) did not help participants
tune in to the model’s action capabilities.
These results support the idea that individuals may be
able to acquire knowledge about another person from the
kinematic patterns produced when the other person moves.
This ability seems to be limited to the case of observing
movements that are functionally related to the task being
judged. That is, kinematic information relating to one class
of body movements appears to be informative about another
non-identical, but functionally similar, class of body
movements. Though the specific force-generation
requirements of squatting and jumping are not identical, the
underlying dynamics of the two tasks are related insomuch
as they both involve the capacity to generate vertical force
by contracting the muscles of the legs. Our findings suggest
that this similarity was sufficient to allow participants to
gain information about the model that is important to
performing a vertical jump without actually observing the
model jump.
General Discussion
The results of Experiment 1 suggest that perceivers rely on
distinct sources of information when perceiving maximum
jump-with-reach height for the self and for another actor.
Self-estimates improved over time, even without feedback
about the accuracy of the estimates, but other-actor
estimates did not. In Experiment 2 it was found that greater
accuracy of the estimates for the other actor could be
achieved by allowing the perceiver to observe the actor
perform a functionally related, but non-identical, action.
Being able to perceive the actions another person is
capable of performing may be a crucial component of the
perception-action processes involved in coordinating social
behavior or in predicting what another person is about to do.
Different theoretical approaches have been proposed to
account for social perception-action processes. Embodied
simulation (e.g., Gallese, 2005) casts the problem as one of
using one’s own action capabilities as a model for
apprehending another person’s actions. The present study
did not provide support for that position, and instead
2466
provided evidence for independence of perception of
another person’s potential actions from perception of
possible actions for oneself (Experiment 1). Experiment 2
provided evidence that making available optical information
about another person’s action capabilities can enhance
perception of what is afforded the other person. That result
is consistent with information-based approaches to social
perception-action (e.g., Ramenzoni et al., 2008). In the
present experiments, however, the perceivers and models
never engaged in any direct form of social interaction. In
future studies it may be interesting to manipulate the degree
of interaction between the perceiver and model, as the
opportunity for genuine interaction may promote a greater
sensitivity of the perceiver to the other person’s action
capabilities. The theoretical move that is now called for is to
develop a comprehensive perspective that encompasses both
the neuro-cognitive and informational constraints on social
perception-action.
Acknowledgments
This research was supported by NSF grant BCS 0716319
(Shockley & Riley).
References
Cesari, P., Formenti, F., & Olivato, P. (2003). A common
perceptual parameter for stair climbing for children,
young and old adults. Human Movement Science, 22, 111124.
Dowling, J. J., & Vamos, L. (1993). Identification of kinetic
and temporal factors related to vertical jump performance.
Journal of Applied Biomechanics, 9, 95-110.
Fajen, B. R. (2005). Perceiving possibilities for action: On
the necessity of calibration and perceptual learning for the
visual guidance of action. Perception, 34, 741-755.
Fajen, B. R. (2007). Affordance-based control of visually
guided action. Ecological Psychology, 19, 383-410.
Gibson, J. J. (1986). The ecological approach to visual
perception. Hillsdale, NJ: Lawrence Erlbaum Associates.
(Original work published 1979).
Guilford, J. P. (1954). Psychometric methods. New York:
McGraw-Hill.
Konczak, J., Meeuwsen, H. J., & Cress, M. E. (1992).
Changing affordances in stair climbing: The perception of
maximum climbability in young and older adults. Journal
of Experimental Psychology: Human Perception and
Performance, 18, 691-697.
Mark, L. S. (1987). Eye-height scaled information about
affordances: A study of sitting and stair climbing. Journal
of Experimental Psychology: Human Perception and
Performance, 13, 360-370.
Mark, L. S., Balliet, J. A., Craver, K. D., Douglas, S. D., &
Fox, T. (1990). What an actor must do in order to
perceive the affordance for sitting. Ecological
Psychology, 2, 325-366.
Oudejans, R. R. D., Michaels, C. F., Bakker, F. C., &
Dolné, M. (1996). The relevance of action in perceiving
affordances: Perception of catchableness of fly balls.
Journal of Experimental Psychology: Human Perception
& Performance, 22, 879-891.
Pepping, G. J., & Li, F.-X. (1997). Perceiving action
boundaries in the volleyball block. In M. A. Schmuckler
& J. M. Kennedy (Eds.), Studies in perception and action
IV (pp. 137-140). Mahwah, NJ: Lawrence Erlbaum
Associates.
Pepping, G. J., & Li, F.-X. (2000). Changing action
capabilities and the perception of affordances. Journal of
Human Movement Studies, 39, 115-140.
Pepping, G. J., & Li, F.-X. (2005). Effects of response mode
on reaction time in the detection of affordances for
overhead reaching. Motor Control, 9, 129-143.
Ramenzoni, V. C. Riley, M. A. Davis, T., & Snyder, J.
(2005). Perceiving whether or not another person can use
a step to reach an object. In H. Heft & K. L. Marsh (Eds.),
Studies in perception and action VIII (pp. 15-18).
Mahwah, NJ: Erlbaum.
Ramenzoni, V. C., Riley, M. A., Shockley, K., & Davis, T.
(2008). An information based approach to action
understanding. Cognition, 106, 1059-1070.
Ramenzoni, V. C., Riley, M. A., Shockley, K., & Davis, T.
(in press). Carrying the height of the world on your
ankles. Quarterly Journal of Experimental Psychology.
Ramenzoni, V. C., Riley, M. A., Davis, T., Shockley, K., &
Armstrong, R. (in press). Tuning in to another person’s
action capabilities: Perceiving from walking kinematics
the height to which one can jump and reach. Journal of
Experimental Psychology: Human Perception and
Performance.
Runeson, S. (1983). On visual perception of dynamic events.
Acta Universitatis Upsaliensis: Studia Psychologica
Upsaliensia (Serial No. 9). (Original work published
1977).
Runeson, S., & Frykholm, G. (1981). Visual perception of
lifted weight. Journal of Experimental Psychology:
Human Perception and Performance, 7, 733-740.
Runeson, S., & Frykholm, G. (1983). Kinematic
specification of dynamics as an informational basis for
person-and-action perception: Expectation, gender
recognition, and deceptive intention. Journal of
Experimental Psychology: General, 112, 585-615.
Runeson, S., Juslin, P., & Olsson, H. (2000). Visual
perception of dynamic properties: Cue heuristics versus
direct perceptual competence. Psychological Review, 107,
525-555.
Sebanz, N., Bekkering, H., & Knoblich, G. (2006). Joint
action: bodies and minds moving together. Trends in
Cognitive Sciences, 10, 70-76.
Stoffregen, T. A., Gorday, K. M., Sheng, Y-Y., & Flynn, S.
B. (1999). Perceiving affordances for another person's
actions. Journal of Experimental Psychology: Human
Perception & Performance, 25, 120-136.
Warren, W. H. (1984). Perceiving affordances: Visual
guidance of stair climbing. Journal of Experimental
Psychology: Human Perception and Performance, 10,
683-703.
2467