S1 (FriAM 1-65) - The Psychonomic Society

Saturday Afternoon Papers 244–250
In several experiments, memory was impaired for new contextual information
presented with an emotional harbinger when the anticipated
negative or positive emotional outcome did not occur. Furthermore,
there was retrograde amnesia for the emotional harbinger itself as well
as forgetting of neutral information following it. This emotional harbinger
effect may help shed light on the mechanisms of disorders involving
impaired emotional expectations, such as anxiety, phobias,
and compulsive gambling.
5:10–5:25 (244)
Quantitative Review of 25 Years of Research on Prospective Memory.
BOB UTTL & KIMBERLY BALTIMORE, Red Deer College—We
rely upon prospective memory proper (ProMP) to bring back to awareness
previously formed plans and intentions at the right place and
time—for example, a plan to buy groceries en route home. ProMP is
distinguished from other subdomains of prospective memory (ProM)
such as vigilance/monitoring and habitual ProM. Our meta-analysis
of several hundred articles accumulated over 25 years of research on
ProM shows that (1) many studies suffer from widespread methodological
problems (e.g., ceiling effects, small sample sizes) and conceptual
confusions (e.g., failure to distinguish between ProM subdomains),
(2) many aspects of ProM have received no or only minimal
attention (e.g., event cued ProMP examined in natural settings), and
(3) the ProM instruction to ongoing task start delay as well as a number
of ProM cues encountered by each participant influence performance
on ProM tasks, supporting the distinction between vigilance/
monitoring and ProMP.
Categorization
Beacon B, Saturday Afternoon, 3:30–5:30
Chaired by Donald A. Schumsky, University of Cincinnati
3:30–3:45 (245)
Response-Time Approach to Contrasting Rule and Exemplar Models
of Classification. ROBERT M. NOSOFSKY & MARIO FIFIC,
Indiana University, Bloomington—We collect response-time data in
tasks of speeded classification to contrast the predictions of serialprocessing
rule-based models, parallel-processing rule-based models,
and exemplar models. The paradigm builds upon the “systems-factorialtechnology”
used by researchers for identifying alternative mentalprocessing
architectures. Initial validation tests show that in conditions
involving separable-dimension stimuli and in which observers
are given explicit instructions for use of a serial rule-based strategy,
the technology recovers the use of this strategy. Ongoing tests contrast
the predictions from the models in situations involving integraldimension
stimuli and in situations in which observers learn category
structures via induction over training exemplars. Beyond contrasting
rule and exemplar models, the research aims to unite mentalarchitecture
and random-walk approaches to modeling choice and response
times.
3:50–4:05 (246)
Similarity and Categorization, Single or Multiple Processes? An
Application of the Reverse Association Test. TAKASHI YAMAUCHI,
Texas A&M University—The reverse association test (Dunn & Kirsner,
1988) is a powerful procedure to assess the link between underlying
cognitive processes and task performance. It offers strong evidence for
the presence of multiple processes/systems underlying two related
tasks. However, this test has not been fully utilized. The test often requires
a large scale single study employing at least 6 independent manipulations.
Furthermore, the statistical procedure to verify reverse association
has not been clearly delineated. This article examines the
relationship between categorization and similarity judgments and presents
practical solutions for these problems. First, a meta-analytic procedure
comparing multiple studies is proposed. Second, a contrast
analysis for the detection of reverse association is illustrated in a case
study. This case study reveals reverse association in categorization and
38
similarity judgments, providing additional evidence for the view that
categorization and similarity rely on separable cognitive processes.
4:10–4:25 (247)
Promoting the Use of Less Likely Category Alternatives in Property
Induction With Uncertain Categories. BRETT K. HAYES &
BEN R. NEWELL, University of New South Wales—Three experiments
examined how people make property inferences about instances
whose category membership is uncertain. Participants were shown
two categories and a novel instance with a feature that indicated the
instance was more likely to belong to one category (target) than to the
other (nontarget). Participants then made categorization decisions and
property inferences about the novel instance. In some conditions property
inferences could only be made by considering both target and
nontarget categories. In other conditions predictions could be based
on both categories or on the target category alone. Like previous studies
(e.g., Murphy & Ross, 1994, 2005), we found that many people
made predictions based on consideration of only the target category.
However, the prevalence of such single-category reasoning was reduced
by highlighting the costs of neglecting nontarget alternatives
and by asking for inferences before categorization decisions. The implications
for process models of uncertain inference are discussed.
4:30–4:45 (248)
Judging Discrete and Continuous Quantity: In Large Sets, Number
Counts. HILARY C. BARTH, Wesleyan University—Recent studies
suggest that interference from automatically extracted irrelevant quantitative
dimensions affects adults’ judgments of discrete and continuous
quantity. Four experiments provide evidence against this idea.
Adults made judgments of the total continuous quantity present in a set
of elements. Continuous amount was not rapidly and automatically extracted;
comparisons of possible models of individual performance
demonstrate that participants arrived at estimates of total continuous
amount through at least four different unconscious computations. A
fourth experiment required participants to make judgments of the number
of elements in a set. Comparisons of possible models of individual
performance show that a single model can explain each participant’s responses,
and that all participants based their judgments on discrete
number. These findings demonstrate large qualitative differences between
adults’ routes to discrete and continuous quantity: the computation
of total continuous quantity is inaccurate and idiosyncratic, unlike
the quick and effective extraction of discrete numerical information.
4:50–5:05 (249)
Absolute Frequency Identification: How a Sequence of Events Can
Affect People’s Judgments. PETKO KUSEV & PETER AYTON,
City University London, PAUL VAN SCHAIK, University of Teesside,
& NICK CHATER, University College London (sponsored by James
Hampton)—Theories of absolute identification and categorization established
over past decades have revealed people’s inability to classify
or judge perceived objects independently of their context. Our series
of experiments shows that judged frequencies of sequentially encountered
stimuli are affected by certain properties of the sequence
configuration: Representations of a category do not only depend on
(1) the number of stimuli/chunks in the sequence (2) the relation of
the current stimulus to the immediately preceding stimulus, and (3) relations
between stimuli further back versus recent stimuli but also on
(4) simple sequence characteristics. Specifically, a first-run effect occurs
whereby people overestimate the frequency of a given category
of event when that category is the first repeated category to occur in
the sequence. While current theories of absolute identification and
categorization cannot account for these results, the simplicity framework,
suggested in this paper, accounts for this and other context effects
on judgment tasks.
5:10–5:25 (250)
Evolutionary Models of Color Categorization Based on Discrimination.
KIMBERLY A. JAMESON, NATALIA L. KOMAROVA, LOUIS