come noisy and less reliable for slow movements.
Thus, making dissimilarity ratings more reliant on
factors other than the velocity profile.
5. Summary of the two experiments
Results of the two experiments can be used to rea-
son what properties of movements are salient in the
recognition of movement. In Experiment 1 it was
seen that movements with distinctly different hand
paths such as those generated by technique MT for
movement (a) and technique EPH for movement
(b) were seen as distinctly different. While this
could have been expected based on casual viewing
the result serves as baseline data and confirmation
of a quantitative method for validating such infor-
mal observations. Experiment 2 discarded the out-
liers obtained in the previous experiment leaving a
set of movements that were not distinctly different
from one another in their hand path and tried to
find what movement properties were used to distin-
guish between the movements. It was found qual-
itatively that the fast movements appeared to be
distinguished from one another based on their ve-
locity profiles. The slow movements appeared to be
distinguished from their velocity profile and another
factor. We conjecture that this additional factor is
related to the posture of the body, but further inves-
tigation is necessary.
6. Conclusions
Of theoretical interest in the interpretation of the
current results is the nature of the difference be-
tween the perception of the fast and slow move-
ments. That there would be a difference is consistent
with results that show that as more time is given
to view a human motion sequence there is an in-
creasing tendency for a cognitive interpretation to
be given to it. In this light we can conjecture that
the sensitivity to differences in velocity profiles is
due to low-level perceptual processing of the move-
ments and that for slow movements the addition of
another dimension in the MDS solution is reflect-
ing some cognitive analysis of the movements, al-
though there are alternatives to this explanation.
For example, it has been argued that the percep-
tion of human movement is mediated not through
motion per se, but rather through the recogni-
tion of discrete postures (Beintema and Lape, 2002,
Giese and Poggio, 2003). Such a mechanism might
be more influential for slow movements.
Further research will probe this question, investi-
gating which features of motion are salient for recog-
nition excluding velocity, and when they become sig-
nificant with respect to the influence of the velocity
profile.
References
Abend, W., Bizzi, E., and Morasso, P. (1982). Hu-
man arm trajectory formation. Brain, 105:331-
348.
Atkeson, C. and Hollerbach, J. (1985). Kinematic
features of unrestrained arm movements. Jour-
nal of Neuroscience, 5:2318-2330.
Atkeson, C. G., Hale, J. G., Pollick, F., Riley, M.,
Kotosaka, S., Schaal, S., Shibata, T., Tevatia,
G., Ude, A., Vijayakumar, S., and Kawato, M.
(2000). Using humanoid robots to study human
behavior. IEEE Intelligent Systems Magazine,
special issue on Humanoids, 15(4).
Beintema, J. and Lape, M. (2002). Perception of bi-
Iogical motion without local image motion. Pro-
ceedings of the national Academy of Sciences,
99:5661-5663.
Bertenthal, B. (1993). Infants’ perception of
biomechacnical motions: Intrinsic image and
knowledge-based based constraints. In Granrud,
C., (Ed.), Carnegie symposium on cognition:
Visual perception and cognition in infancy,
pages 175-214, Hillsdale, NJ. Erlbaum.
Bizzi, E., Polit, A., and Morasso, P. (1976). Mech-
anisms underlying the achievement of final head
position. Journal of Neurophysiology, 39:434-
444.
Feldman, A. (1966). Functional tuning of the ner-
vous system with control of movement or main-
tenance of a steady posture -iii. mechanographic
analysis of the execution by man of the simplest
motor tasks. Biophysics, 11:766-775.
Flash, T. (1983). Organizing principles underlying
the formation of hand trajectories. Ph.D. thesis.
Harvard∕MIT Division of Health Sciences and
Technology.
Flash, T. (1987). The control of hand equilibrium
trajectories in multi-joint arm movements. Bio-
logical Cybernetics, 57:257-274.
Flash, T. and Hogan, N. (1985). The co-ordination
of arm movements: An experimentally con-
firmed mathematical model. Journal of Neu-
roscience, 5:1688-1703.
Fox, R. and McDaniel, C. (1982). The perception
of biological motion by human infants. Science,
218:486-487.
Giese, M. and Poggio, T. (2003). Neural mech-
anisms for the recognition of biological move-
ments. Nature Reviews Neuroscience, 4:179-
192.