PERCEPTION DE L'ESPACE 291
lancement ~t distance. Et aussi avec le choc A distance on peut
obtenir un effet cauml tr~s marqu6 en aceentuant la hi6rarchie des
vitesses.
Nos exlxSriences montrent que l'orientation des deux mouvements,
de l'agent et du patient, clans la mSme direction ou dans des
directions peu diver$cntes est *galement une condition t r~
favorable, mais pas absolument n~:essaire, car en compensant
l'absence de cette condition on peut aussi obtenir une impression
causale lorsque le mouvement du patient se d6ve- loppe clans une
direction diam~tralement oppost.~e A celle du mouvement de
l'agent.
De cette faqon on a pu produire des effets paradoxaux de
lancement invers6 /~ distance; tandis qut: d'autres situations, au
lieu d'etre v~cues comme formes de lancement, r~alisent sur le plan
ph~nom6nal des im- pressions d'attraction active. On a obtenu aussi
ce meme ph6nom~ne d'attraction dans des situations qui ne sont pas
assimilables au lancement mais semblent pr~cnter une affinit6 avee
l'effet entrainement.
Du point de rue th6orique ces ph~nom~nes semblent rentrer, sans
effort, dans 1 cadre du sch6ma interpr~tatif du Pr. Michotte, fond6
sur l'aml~ liation du mouvement.
A STUDY OF STABILITY OF DISCRIMINATION IN A VISUAL
MATCHING TEST
BY
FORREST L. DLMMICK
(U.S. Naval Medical Research laboratory, New London, Conn. )
Since color matching is based upon discrimination of
differences, the scatter ofa succes~fion of repeated matches will
reflect the normal distribution of discrimination judgments. To
evaluate this relationship, each one of a group of six subjects
repeated a test of 40 matches ten times. The matching colors
consisted of ten saturation steps in each one of the principal
hues, red, yellow, green, and blue. Differences between adjacent
steps in the series were of a size that precluded correct matching
of all 40 samples by any subject except in very rare cases.
It was found that (a) the group was relatively homogeneous with
mean scores rangi,lg from 61-70, (b) th~ standard deviations were
small, ranging from 4- 4 to 4- 8, (c) the data i~Micated that more
precise measures may
292 ~u~ 10
be obtained by several successive tests of any given subject,
and (d) the judL, ments were normally distributed.
The distributions of errors for each of the four hues were
essentially the same. The standard deviation of each distribution
was computed. T!~ese were arbitrary units differing in the case of
each hue. The standard deviations of the red and the yellow hues
were compared with the just- noticeable-differences of saturation
for comparable colors from a previous study. It is interesting that
these two just-noticeable-differences had almost identical ratios
to the standard deviations of the present analysis.
Observation of techniques used in matching and reports from
subjects indicated that a "match" was never a simple and direct
judgment of equality. It was instead an inference from, or an
evaluation of, a number of discrimina- tions or judgments of
difference. For most observers, the possibilities of match were
quickly reduced to two or three chips that were not readily
discriminated as different. The final choice tended to reflect the
humor of times each pair of chips was discriminated. Single
matches, obviously, were not fixed. This accounts for the greater
stability of scores with repeated testing.
SOME CONTRIBIY(IONS TO THE STUDY OF COMPLEX
PERCEPTIONS
BY
ENZO SPM.TRO
( Universint Cattolica. Milano)
Perceptive structures are often influenced by non-perceptive
structures. However the researches about the subjective males of
evaluation seem to find that these non-perceptive structures can be
employed for a "measure- ment" of the perceptive structures.
Otherwise the uncertain income is very important in the forming
of perceptive structures: this uncertain income is the
"aspettativa" that a perceptive event may happen; the researches
about the subjective probability seem to have proved all 'this.
In this research the aether has been searching a rational point
of view in the study of the problem of the importance that some
evaluative structures have in the forming of a "complex
perception".
That, which was studied, of a linear mathematical function is
the founda-
