When ones catches a free-falling object, the brain manages to made prediction
of how long does it takes to the ball to touch the hand. Lacquaniti and Maioli
(1989) have shown that this prediction is characterised not only by an anticipatory
activation (about 100 ms before impact) of EMG signals recorded on flexor muscles
of the elbow, but also by an intensity of activation proportional to the impact
of the ball in the hand.
Their experiment was based on a 3x3 factorial design : 3 different heights and
3 balls of different mass but identical size. As subjects were able to see the
height of release, one could think that visual information available before
the fall had been slanted. Indeed, it can be supposed that central nervous system
(CNS) used those information to compute an appropriate muscular response relative
to the expected perturbation of the arm posture caused by the impact.
In order to test the hypothesis that SNC made a prediction based on information
either before (height) or after releasing (trajectory) the ball itself, we performed
an experimental protocol similar to Lacquaniti and Maioli (1989) but with new
devices.
Six volunteers were asked to catch a Silicon® ball (? : 9cm / 400 g) projected threw the floor by a launcher which opening was at 1.6 m high above the hand of the subjects. We have used 3 initial speeds of release (1, 2 and 3 m/s) which permit us to simulate different heights of falling. Kinematic data of the arm (from reflected markers positioned on arm's joints) and EMG data of biceps brahcii were synchronised and recorded by mean of an optokinetic system (Kinelite® 4-TV) with a sample frequency of 200 and 800 Hz, respectively.
The same pattern of muscular activity has been observed. It is to say, anticipatory activity of biceps brachii began before ball impact and intensity of muscle activation was proportional to momentum of the ball (i.e. mass x velocity, that become impact variation if expressed relative to time).
As the height of the release does not change, it would seem that visual information
are use in that task. Nevertheless, different models (Lee 1976; Savelsbergh
and al. 1991, 1992) based only on visual cues or variables fail in predicting
our results.
As a matter of fact, the ball does not fall at constant velocity and the trajectory
is not perpendicular to the front plan of the subject. We hypothesis that integration
of visual information might be coupled to proprioceptive and gravitational information
together in an internal model which allows to anticipate the perturbation due
to ball impact in the hand and produce the right muscular activity at the right
relative time.
Lacquaniti F. & Maioli C. (1989). The role of preparation in tuning anticipatory and reflex during catching. J. Neurosci., 9: 134-148.