harnad@phoenix.Princeton.EDU (S. R. Harnad) (04/10/90)
*** PSYCOLOQUY: Sponsored on an experimental basis by the Science
Directorate of the American Psychological Association 202/955-7653 ***
C.P.C. - CAHIERS DE PSYCHOLOGIE COGNITIVE
EUROPEAN BULLETIN OF COGNITIVE PSYCHOLOGY
VOLUME 10 NUMBER 1 FEBRUARY 1990
From: Francoise Joubaud
Address: IBHOP, Traverse Charles Susini, 13388 Marseille Cedex 13
France - Tel: (33) 91.66.00.69 - Fax: (33) 91.66.00.69 -
E-mail: CPC at FRMOP11
An analysis of the association rates between numbers from 1 to 20.
Jean Costermans
Abstract:To each number from 1 to 20, 196 subjects were asked to
associate three other numbers chosen within the same range. The
analysis of the frequency tables allowed us to define 21 types of
relationships, distributed into four classes: neighbourhood on the
magnitude scale (e.g., N+1), symmetry relative to 10 (e.g., N-10),
simple ratios (e.g., N/2), and common membership of some typical
number families (e.g., primes). The weight of each factor was
estimated, and possible interactions were examined when two or
more factors were involved in a single number pair. Some organiza-
tional principles were proposed. When used for deriving predictions
about response frequencies, they yield a fairly good adjustment
between predicted and observed association rates. (pp. 3-21).
Intergroup attributions for success and failure: Group-serving
bias and group-serving causal schemata
Miles Hewstone, Lynn Gale, and Nicola Purkhardt
Abstract: Research on intergroup attributions for success and
failure has reported the effects of social categorization and
outcome for each causal attribution separately, but has ignored
the relative importance of the four conventional causes. A complete
method of analysis is outlined, which investigates three factors
- categorization (ingroup/outgroup) x outcome (success/failure) x
attribution type (ability/effort/luck/task) - and then tests
relevant contrasts. In Study 1 this method was used to re-analyse
data and revealed ingroup-enhancing and ingroup-protecting biases.
It also highlighted the different configuration of causes used
to explain ingroup and outgroup failure, but not success. Multi-
dimensional scaling analyses suggested, however, that different
causal schemata might be used to explain ingroup and group
success. Study 2 partially replicated these findings using
unstructured responses, and provided some evidence of single- vs.
multiple-cause schemata for ingroup and outgroup success,
respectively, supported by multidimensional scaling analysis.
Results from the two studies are discussed in terms of the need
for methodological diversity in investigating ingroup-serving
biases in causal attribution. (pp. 23-44).
An experiment in learning logical negation by 6- and 7-year-old
children.
Claude Bastien, Aline Pelissier, and Alain Tete
Abstract: The issue of "context effects" in problem solving
was addressed by conducting a learning experiment using a
complementation task with 6- and 7-year-old children. The
procedure consisted of leading those subjects who produced
partial complementation responses to focus on the relevant
properties of the remaining objects. The results indicate that
with this procedure, 40% of the subjects learned in two trials.
The discussion deals with the nature of such learning. The point
of view defended is that the problem solving task is far from
depending on the situational context, thought simply to modulate
it, but on the contrary, is the origin of this context, whose
limits it defines and whose meaning it determines.(pp. 45-63)
Complementation in children: Negation of objects of negation
of properties?
Annie Desprels-Fraysse
Abstract: Three experiments were carried out to analyze the way
children deal with different complementation tasks between the
ages of six and eight. The identifying power of the properties
of the objects used was varied. The hypothesis that children can
negate identifying properties (note of these are fish, birds,
insects) more easily than accidental properties (none of these
are square, red, dotted) was validated. This relative facilitation
may be a source of developmental progress due to the procedural
transposition that it initiates. (pp. 65-78)
Times series: A tool for analyzing complex cognitive activities.
Application to the study of text revising strategies.
Francois Guercin, Jean-Yves Roussey, and Annie Piolat
Abstract: This article first presents a formal approach to the
study of time series. This approach led us to define time series
as a function of the temporal organization of complex activities
found in psychology, and then to characterize the descriptive
statistical processing in terms of the selection and comparison
of sub-series. Five formal operations suffice to account for
sub-series selection. Combinations of these five operations can
be expressed as formulas to represent specific hypothesis. All
of the possible operation combinations are incorporated into a
computer program. Then the use of time series is illustrated by
an application inthe field of text revision. In particular, an
example is given of how the distance between time series can be
used to measure the difference between a subject's production
and simulated productions in application of one of six strategies
based on a few functional principles. In a more general perspective,
this article discusses how time series can be utilized as a tool for
the investigation of the step-by-step procedures employed by
subjects as they execute complex tasks. (pp. 79-110)
--
Stevan Harnad Department of Psychology Princeton University
harnad@clarity.princeton.edu srh@flash.bellcore.com
harnad@elbereth.rutgers.edu harnad@pucc.bitnet (609)-921-7771