The name is absent

GOUTEUX, THINUS-BLANC, AND VAUCLAIR

Table 2

Data Analysis of the 10 First Trials for Each Experiment and
for the Subjects, Indicating Whether They Searched at the

Correct Location, Made a Rotational Error, or Searched

in the Geometrically Inappropriate Corners

Box

Note. C = correct; R = rotational; N & F = near and far misses.

Experiment 2

In Experiment 2, we tested monkeys' ability to reorient along
both the geometric shape and nongeometric properties of the
experimental room, by using a local cue (colored wall) as in
Hermer and Spelke's (1994) experiment.

Method

Subjects. The same subjects as in Experiment 1 were used. A delay
of 1 day occurred between Experiment 1 and Experiment 2.

Apparatus. The apparatus was the same as in Experiment 1, but one of
the small walls was completely covered by a blue panel (blue-wall condi-
tion). This landmark always occupied the same side with respect to the
rewarded box (see Figure 2).

Procedure. The same procedure as in Experiment 1 was used, but a
control phase was added. The control phase (10 trials at the end of the 50
experimental trials) consisted of making a 180° transfer of the cue (blue
wall) and of the rewarded box. The aim of that "rotation" was to check that
the only information used by the monkeys to get reoriented was the blue
wall in addition to the geometry of the experimental environment. Indeed,
if during the 50 experimental trials, an uncontrolled cue (from inside or
outside the apparatus) served as a landmark and guided the subject to the
reward box, then the rotation of the landmark would have eliminated the
use of such a cue.

Results

Table 3 presents the number of first searches performed by the 3
subjects during the 50 test trials. For each monkey, the data of the
test were subjected to a chi-square one-sample test by which we
compared the observed distribution to the theoretical frequency of
an equal distribution in the geometrically appropriate and geomet-
rically inappropriate above-mentioned categories (i.e., 50% of
chance for each one). These results always reached a statistically
significant level, Orcas, X2(1, N = 50) = 32.0; Krill, X2(1, N = 50)
= 46.1; Crevet, X2(1, N = 50) = 46.1; all ps < .001. Another
chi-square one-sample test computed on the data obtained in the
two corners of the observed geometrically appropriate category
compared with an equal frequency of distributions of the searches
to these two corners (i.e., 50% of chance for each one), indicated
that the numbers of visits to Corner C and to Corner R were
statistically different, Orcas, A,2 (1, N = 45) = 18.68, p < .001;
Krill, X2(1, N = 49) = 9.00, p < .005; Crevet, X2(1, N = 49) =
22.22; p < .001. The monkeys mainly chose the correct corner.

For Experiment 2, the data of each subject during the entire
experimental session (50 trials) are very similar to those observed
during the first 10 trials (see Table 2, Experiment 2). Thus, we can
conclude that no obvious improvement of the monkeys' perfor-
mance takes place during the experimental procedure, logistic
regression: Orcas, X(1, N = 60) = 0.05; Krill, X2(l, N = 60) _
0.01; Crevet, X2(1, N = 60) = 4.70; p > .05.

During the virtual rotation control session, all the subjects
reacted in the same way. On average, 84% of their first choices
during the 10 trials were directed to the rewarded box (noted "C";
90% for Subject Orcas, 60% for Subject Krill, and 99% for Subject
Crevet). The geometrically equivalent corner (Corner R) received
an average of 13% of the first visits (0% for Orcas, 30% for Krill,
and 1% for Crevet). Finally, the geometrically inappropriate cor-
ners (noted "N" and "F„) received an average of 3% (0% for
Orcas, 10% for Krill, and 0% for Crevet) of the first visits. This

Table 3

Number of Trials (Out of 50) During Experiment 2 for the
Subjects as a Function of Their Search Location (Correct,
Rotational, Geometrically Inappropriate Corners)

Box

Note. C = correct; R = rotational; N = near misses; F = far misses.

<<   <   3   4   5   6   7   8   9   >   >>