SATURDAY, November 2, Grand Ballroom F, East Tower Ballroom Level
8:00 am- 9:30 am Psychonomic Society Probability
Judgment Session
(see abstracts, p. 8)
9:40 am-12:00 pm Psychonomic Society JDM Session I
(see abstracts, p. 8)
12:00 pm- 1:30 pm Psychonomic Society Poster Session
(in Columbus Hall, East Tower
Ballroom Level)
12:00 pm-12:30 pm SOCIETY FOR JUDGMENT AND DECISION MAKING
Announcements and Registration
1:30 pm- 3:40 pm Psychonomic Society JDM Session II
(see abstracts, p. 9)
3:50 pm- 5:35 pm Psychonomic Society JDM Session III
(see abstracts, p. 9)
5:35 pm- 6:00 pm SOCIETY FOR JUDGMENT AND DECISION MAKING
Announcements and Registration
6:00 pm- 7:30 pm Psychonomic Society JDM Poster Session
(including JDM topics; in Columbus
Hall, East Tower Ballroom Level)
8:00 pm- ??? SOCIETY for JDM SOCIAL EVENT:
Second City Comedy Club Show
Don't miss our trip to the hilarious Second
City Comedy Club! Space is limited, so
reserve your seat now. The cost of tickets
is $15.50. (See conference registration
form on page 26.)
SUNDAY, November 3, ROOMS TBA
8:00 am- 8:30 am Registration
8:30 am-10:00 am SYMPOSIUM
HONORING THE INFLUENCE OF AMOS TVERSKY: A
MEMORIAM BY HIS STUDENTS, organized by Valerie
Reyna (Univ of Arizona) & Baruch Fischhoff (Carnegie
Mellon)
One of the many ways in which Amos Tversky has
profoundly influenced our field is through his
mentoring of students. In this symposium, we honor
the memory of Amos Tversky by describing this
influence, while highlighting different aspects of
Amos' life and work.
Opening remarks (Valerie Reyna & Baruch Fischhoff)
Craig Fox (Duke Univ)
A Belief-based Approach to Decision Under
Uncertainty
Derek Koehler (Univ of Waterloo) & Lyle Brenner
(UCLA)
The Enhancement Effect in Probability
Judgment
Yuval Rottenstreich (Calif. Inst. of Tech.)
Separate and Joint Assessments of Preference
and Belief
Richard Gonzalez (Univ of Washington)
On the Weighting Function of Prospect
Theory
Closing remarks (Valerie Reyna & Baruch Fischhoff)
10:00 am-10:30 am Break (with continuing registration)
10:30 am-12:10 pm
PARALLEL SESSION A
Moderator: Alan Schwartz (UC-Berkeley)
10:30 am-10:50 am Karl Teigen (Univ of Troms,
Norway)
When the Unreal Is More Likely Than the
Real: Post hoc Probabilities and
Counterfactual Closeness
10:55 am-11:15 am Terry Connolly, Lisa Ordonez, &
Richard Coughlan (Univ. of Arizona)
Regret and Responsibility in the Evaluation
of Decision Outcomes
11:20 am-11:40 am Ravi Dhar (Yale) & Klaus
Wertenbroch (Duke)
Hedonic Loss Aversion and the Behavioral
Economics of Pleasure
11:45 am-12:05 pm Jonathan Baron (Univ of
Pennsylvania), Mark Spranca (Rand
Corp.), Julia Kalmanson (Univ of
Penn.), & Julie Irwin (NYU)
Protected Values
PARALLEL SESSION B
Moderator: TBA
10:30 am-10:50 am Alan Cooke (UC-Berkeley)
Preferences with Missing Information:
Exploring the Inference Process
10:55 am-11:15 am Kathleen Mosier (NASA Ames),
Linda Skitka & Mark Burdick (Univ
of Illinois at Chicago), & Susan
Heers (Monterey Technologies)
Decision Making in a High-Tech World:
Automation Bias and Countermeasures
11:20 am-11:40 am Peter Juslin & Henrik Olsson
(Uppsala Univ, Sweden)
Thurstonian and Brunswikian Origins of
Uncertainty in Judgment: A Sampling
Model of Confidence in Sensory
Discrimination
11:45 am-12:05 pm J. Frank Yates & Paul Estin (Univ of
Michigan)
Training Good Judgment
12:10 pm- 1:00 pm Lunch (Everyone's on their own.)
1:00 pm- 2:15 pm POSTER SESSION #1 (in Columbus
Hall, East Tower Ballroom Level)
(see abstracts, pp. 10-18)
2:15 pm- 2:45 pm EINHORN AWARD
Presented by Donald Kleinmuntz
2:45 pm- 3:45 pm Invited Speaker: Vernon Smith
Game Theory and Reciprocity in Extensive Form
Games
Vernon L. Smith is Regents' Professor of
Economics at the University of Arizona. Known
as the father of experimental market economics,
he is the founder of the University of Arizona's
Economic Science Laboratory, a member of the
National Academy of Sciences, and winner of the
1995 Adam Smith Award.
3:45 pm- 4:15 pm Break
4:15 pm- 5:45 pm
PARALLEL SESSION A
SYMPOSIUM: Naturalistic Decision Making
Chair: Rebecca Pliske (Klein Associates)
Participants:
Gary Klein (Klein Associates)
What Is the Naturalistic Decision Making
Framework All About?
Lee Beach (Univ of Arizona)
Image Theory: Decisions in the Workplace
Judith Orasanu (NASA Ames)
Naturalistic Decision Making in Aviation:
A Dynamic and Complex Domain
Jim Shanteau (Kansas State)
NDM vs. JDM: Cooperation or
Competition?
PARALLEL SESSION B
SYMPOSIUM: Models of Satisficing Inference
Chair: Gerd Gigerenzer (Max Planck Inst., Germany)
Participants:
Gerd Gigerenzer
Introducing Satisficing Models of Inference
and How They Affect Our Notions of
Sound Reasoning and Rationality
Jean Czerlinski (Max Planck Inst.)
How Accurate and Robust Are Fast and
Frugal Algorithms for Making Inferences
in Real-World Environments?
Daniel Goldstein (Max Planck Inst.)
Recognition as a Fundamental Heuristic for
Inference: How Organisms Exploit Their
Own Lack of Knowledge to Make Fast and
Accurate Inferences
Ralph Hertwig (Max Planck Inst.)
A Satisficing Mechanism of Memory
Reconstruction and How It Accounts for
Hindsight Bias
5:45 pm- 7:00 pm POSTER SESSION #2 (in Columbus
Hall, East Tower Ballroom Level)
(see abstracts, pp. 18-25)
9:00 pm-10:30 pm GRADUATE STUDENT SYMPOSIUM and social hour
(in Toronto Room, West Tower Ballroom Level)
Organizer: Barbara DeFilippo (Univ of Oregon)
Participants:
Barbara DeFilippo
Psychology and the Learning Community
Model
Maurice Schweitzer (Univ Of Miami)
No Good Deed Goes Unpunished: Things
Not to Do Well as an Assistant Professor
Mark Spranca (Rand Corp.)
Working at a Think Tank as a Behavioral
Scientist
Judith Orasanu (NASA Ames)
Fight or Flight, the Government and
Aviation Psychology
Paul Slovic (Univ of Oregon, Decision Research)
TBA
MONDAY, November 3, Rooms TBA
7:40 am- 8:55 am Breakfast and BUSINESS MEETING
8:55 am- 9:30 am Charles Gettys, Michael Dougherty, & Eve
Ogden (Univ of Oklahoma)
THE PIECES PUT TOGETHER: An Umbrella
Theory for Likelihood Judgments Called
MINERVA3
9:35 am-10:35 am Invited Speaker: CASS SUNSTEIN
WHY (AND WHAT) COURTS DON'T DECIDE
Professor Cass Sunstein is Karl N.
Llewellyn Distinguished Service Professor
of Jurisprudence, University of Chicago
Law School and Department of Political
Science, and Co-Director, Center on
Constitutionalism in Eastern Europe. He is
the author most recently of "Legal
Reasoning and Political Conflict", and is a
frequent contributor to "The New
Republic" and "The New York Times
Book Review." He is now working on
several topics at the intersection of law,
economics, and psychology.
10:35 am-11:00 am Break
11:00 am-12:15 pm
PARALLEL SESSION A
SYMPOSIUM: Effects of Task, Context, and
Processing Goals on the Construction
of Preferences in Judgment and
Choice
Chair: Gregory Fischer (Duke)
Participants:
Alan Cooke (UC-Berkeley) & Barbara Mellers (Ohio
State)
New Ways to Reverse Preferences
Ziv Carmon (Duke) & Itamar Simonson (Stanford)
Consumer Assessments of Value in Choice
and in Matching
Gregory Fischer, Dan Ariely, Ziv Carmon, & Gal
Zauberman (Duke)
Goal-based Construction of Preferences
Discussant: Baruch Fischhoff (Carnegie Mellon)
PARALLEL SESSION B
TEACHING SYMPOSIUM: JDM for Managers: Learning
by Doing
Organizers: Jay Russo (Cornell) & Paul Schoemaker
How does teaching JDM to managers differ from
teaching it to college students? Managers care about
skills not credentials; they want "take aways" with
immediate value; and they learn better from activities
than from lectures. This session will illustrate
activity-based teaching.
12:15 pm- 1:45 pm Lunch and PRESIDENTIAL ADDRESS
Barbara Mellers (Ohio State)
EMOTIONAL RESPONSES TO RISKY CHOICE
1:45 pm- 3:35 pm
PARALLEL SESSION A
Moderator: TBA
1:45 pm- 2:05 pm David Fetherstonhaugh (Stanford),
Paul Slovic (Univ of Oregon), Steven
Johnson (Decision Research), &
James Friedrich (Willamette Univ)
Insensitivity to the Value of Human Life: A
Study of Psychophysical Numbing
2:10 pm- 2:30 pm Oswald Huber (Univ of Fribourg,
Switzerland)
Win-rate and Cost of Protection in a
Multistage Investment Task: Intra- vs.
Inter-individual Variation
2:35 pm- 2:55 pm Maurice Schweitzer & Leslie
Gomberg (Univ of Miami)
The Impact of Alcohol on Negotiated
Outcomes: Experimental Evidence
3:00 pm- 3:30 pm Jonathan Leland (National Science
Foundation)
Decision Making by Similarity: Toward a
Theory of Anomalies
PARALLEL SESSION B
Moderator: Claudia Gonzalez-Vallejo (SUNY-Albany)
1:45 pm- 2:05 pm Amnon Rapoport (Univ of Arizona) &
Darryl Seale (Univ of Alabama at
Huntsville)
Coordination Success and Adaptive Learning
in Market Entry Games with Asymmetric
Players
2:10 pm- 2:30 pm Ido Erev, Gilly Naheer, & Eran Elias
(Technion-Israel Inst. Of Tech.)
On the Decisions of Basketball Referees and
the Advantage of Left-Side Offense Outside
the USA
2:35 pm- 2:55 pm Lehman Benson III & Lee Roy Beach
(Univ of Arizona)
The Effects of Time Constraints on the
Pre-choice Screening of Decision Options
3:00 pm- 3:30 pm George Loewenstein, Drazen Prelec,
& Catherine Shatto (Carnegie Mellon)
A Visceral Account of Impulsivity
Society for Judgment and Decision Making 1996 Poster Sessions
Poster Session #1: Sunday Nov. 3, 1:00 - 2:30 pm
P1-1
Enhancement of Insight
Yi-Han Kung & Hal R. Arkes (Department of Psychology, Ohio University)
Undergraduates made judgments concerning the desirability of various roommates, each of which was described by eight characteristics (Reilly & Doherty, 1992). The experimental group subjects indicated after each trial which cues were important in influencing the judgments on that trial. All subjects at the end of the task provided subjective weights concerning the overall importance of each cue. The relation between these final subjective weights and beta weights was much higher in the experimental group than in the control group, whose subjects provided no cue importance information after each trial.
P1-2
Tradeoffs in Risk Attributes: The Joint Effects of Dimension Preference and Vagueness
Kristine M. Kuhn, David V. Budescu, Adrian Rantilla, James Hershey, & Karen Kramer (Department of Psychology, University of Illinois)
Eighty-seven subjects evaluated hypothetical health risks which varied in terms of the probability of loss, amount of loss, and the precision with which each dimension was specified. In making pairwise choices, most subjects were more concerned with loss amounts than probabilities, and they typically held congruent attitudes toward vagueness on both dimensions. Dimension preference was generally a better predictor of choice than vagueness attitudes, except for pairs which did not allow a choice based on the superiority of one dimension. Ratings of riskiness and favorability were marginally affected by vagueness; vagueness seeking was somewhat less common with ratings than choice.
P1-3
A Generalization of Ellsberg's Paradox to Vague-Vague Cases
Karen M. Kramer & David V. Budescu (Department of Psychology, University of Illinois at Urbana-Champaign)
In most previous generalizations of Ellsberg's 2-color paradox only precise-vague cases were considered. This empirical study examines the existence of the paradoxical pattern of choices in vague-vague cases, i.e. where neither probability is specified precisely. The paradox was found to be prevalent with the vague-vague cases. It occurred more often when differences between ranges of vagueness were large and occurred less often with extreme midpoints. Consistent choices were more frequent than indifferences for extreme midpoints. There was weak evidence that subjects could be classified into distinct groups according to their choice patterns.
P1-4
Alternative Consistency Measures For Ratio-Scaled Judgement Matrices
Bradley D. Crouch & David V. Budescu (Psychology Department, University of Illinois)
The notion of judgment consistency is central to ratio scaling techniques such as Saaty's (1978) Analytic Hierarchy Process. Unfortunately, current hypothesis tests of inconsistency are 1) dependent upon arbitrary assumptions about the uniformity of random pairwise ratio judgments and 2) inextricably linked to one of the many specific scaling algorithms available. Via Monte Carlo simulations, it is shown that these hypothesis tests are not accurate when alternative but equally plausible distributions of these judgments are assumed. In addition, two alternative approaches are proposed and assessed. The first index is computed directly from the matrix of pairwise comparisons thus it is independent of the particular method used to fit scale values. Second, a bootstrapping approach is outlined that uses samples from the observed distribution of judgments for reference rather than an unjustified uniform distribution.
P1-5
Effect of Protocol of Play and Social Orientation in Sequential Resource Dilemmas
Winton Au, David V. Budescu, & Xiao-Ping Chen (University of Illinois & Hong Kong University of Science and Technology)
Individuals' requests in a resource dilemma were compared under four protocols of play. In the simultaneous protocol subjects made their requests simultaneously. Subjects made requests sequentially in other protocols. In the "Sequential" protocol (SEQ) subjects knew their position and the total previous requests; In the "Positional" (POS) and "Cumulative" (CUM) protocols, subjects knew only their position, and the previous requests, respectively. We found a "position effect" in all sequential protocols: requests and position are negatively correlated. The effect was identical in SEQ and CUM, but was weakest in POS. We conclude that although temporal priority can induce the position effect, differential information regarding others' behaviors is more pivotal in determining behavior.
P1-6
A Probabilistic Model of Criticality in a Sequential Public Good Dilemma
Winton Au, S.S. Kormita, & Xiao-Ping Chen (University of Illinois & Hong Kong University of Science & Technology)
A special case of a social dilemma is the provision of a public good. A public good (PG) is a commodity or service made available to all members of a group, and its provision depends on the voluntary contribution of its members. Once provided, all members can enjoy the members of a group, and its provision depends on the voluntary contribution of its members. Once provided, all members can enjoy the benefits of the PG, regardless of whether they contributed or not; hence, there is a temptation to "free-ride" (not contribute) in the hope that others will contribute. Rapoport (1987) showed that an important factor that affects cooperation (contribution) in a PG dilemma is the extent to which a group member is critical in providing it. Erev and Rapoport (1990) proposed a game-theoretical model that yields deterministic predictions about the effects of criticality on cooperation in public good dilemmas. Based on their work and research by Chen, Au, and Komorita (1996), we propose a probabilistic model of criticality. The model was tested and found to fit the empirical data, but with one exception. Explanations are offered for the discrepancy and future research is discussed.
P1-7
What Price For Fairness: A Bargaining Study
Xiao-Ping Chen & Rami Zwick (Dept. of Management of Organizations, Hong Kong University of Science & Technology)
We tested the hypothesis that "fairness" has a price and the higher the price the lower the "demand" for it. Students negotiated in pairs over the division of $50 using a finite horizon fixed cost (per rejection) alternating offer rule. Each pair consisted of a high-cost and a low-cost bargainer. In accordance with the hypothesis, the degree to which the high cost bargainers persisted on seeking fairness was a function of how much it exploited their strategic advantage depended upon their own costs of rejection.
P1-8
Kantian Moral Reasoning as a Guide to Decision Making: A Synthesis of Empirical Findings from Behavioral Decision Research, Theoretical Advances in Evolutionary Psychology, and Philosophical Inquiry into Ethics
Steve Freeman (Sloan School of Management, MIT)
In this paper I argue that rational choice theory is flawed as a theoretical and prescriptive model for decision making; that is rests upon an evolutionary argument that is seriously incomplete. A rigorous analysis of evolutionary processes suggests little reason to expect people to have well-developed consequential reasoning abilities or proclivity; rather, such analysis predicts important advantages in being able to reason about roles, rules, and situations. I offer Kantian moral reasoning as an alternative basis for decision making models.
P1-9
A Failure to Support Support Theory
Steve E. Edgell, Robert M. Row, Woo-Kyoung Ahn, & Loranel Graham (Department of Psychology, University of Louisville & Yale University)
Tversky and Koehler's (1994) support theory predicts that probability estimates of an exhaustive set of explicit hypotheses will sum to one (plus or minus error). Subjects estimated the probabilities of outcomes defined by a spinner. We found the sum to consistently exceed one, which is what is usually found with implicit hypotheses. We also show a class of subjective probability estimates that support theory's fundamental axiom cannot fit.
P1-10
Unpacking Racing and Football Odds: Are Four Legs More Incoherent Than Two?
Peter Ayton (Psychology Department, City University, London)
Support theory (Tversky and Koehler, 1994) suggests that different descriptions of the same event can lead to different evoked subjective probabilities. For football matches English bookmakers quote odds so one can bet, generally, on victory, draw or loss for each team; one can also bet more specifically on the actual score of the game or the result at half-time and full-time. These latter possibilities represent different unpackings of general hypotheses that each team will win. In a comparison with odds for horse races I find incoherences in bookmakers odds - some predicted and some not predicted by Support theory.
P1-11
A Study of Food Risk Perception and its Relation to Purchasing Behaviour
Fergus Bolger, Alistair Sutcliffe, Peter Ayton, & Paul Sparks (Centre for HCI Design & Psych Dept., City University & Institute of Food Research)
Studies of risk perception have found two main factors by which people appraise risk: the degree of "knowledge" regarding a hazard and its "severity". However, it is not known whether these dimensions play a role in choice behaviour. In the present study information search, preferences and choices are recorded during a simulated shopping trip based on an information-board design. This method, plus verbal protocols and post-experimental interviews, are used to capture the policies people follow when buying food. The degree to which policies can be represented as linear models will be investigated. The relationship between individual differences in risk orientation and purchasing behaviour will also be examined.
P1-12
Arguments, Uncertainty, and Non-numerical Risk Communication
David Hardman, Peter Ayton, John Fox, & Alistair McClelland (City University, Imperial Cancer Research Fund, & University College, London)
One way of assessing risk in the absence of reliable statistical information is to construct arguments for and against the likelihood of negative outcomes. The resulting judgment may then be expressed using linguistic uncertainty terms. But how do people interpret such terms? In a series of experiments we examined the way in which terms such as "probable" and "plausible" were used in relation to arguments about realistic topics. Results suggest that such terms have a fairly reliable structure characterised by two main underlying dimensions: Polarity and Uncertainty. This work is motivated by the need to develop new methods for communicating information about risk.
P1-13
Resource-Allocation Behavior in Common Two- and Three- Dimensional Settings
Harvey J. Langholtz, Christopher T. Ball, Barron M. Sopchak, & Jacqueline M. Auble (The College of William & Mary & University of Melbourne, Australia)
How do people allocate resources? How will the structure or complexity of the resource-allocation problem affect people's resource-allocation decisions? In three earlier studies Langholtz, Gettys, and Foote (1993, 1994, and 1995) examined resource-allocation behavior where participants solved technical problems and had the opportunity to allocate resources on a continuous scale. In the present study participants solved a resource-allocation problem that is universal and common, and allocated resources on a discreet scale. Additionally, the effects of task structure were included as we examined behavior in two- and three-dimensional problems. The data support the generalization of earlier findings to include nontechnical problems on a discreet scale in three dimensions.
P1-14
A Protocol Analysis of Resource-Allocation Decisions
Christopher T. Ball, Harvey J. Langholtz, Jacqueline M. Auble, & Baron M. Sopchak (Department of Psychology, The College of William and Mary and University of Melbourne, Australia)
Deciding how to allocate resources is a common dilemma for most decision makers. Managers can make use of computer software for providing optimal solutions, but how do others approach such decisions? The college students who participated in this experiment had to allocate a fixed amount of time and money for meals in a week, and provide verbal protocols while making these decisions. The encoded protocols highlighted different strategies for allocating resources, and the performance using each strategy was compared with the optimal solution. Most strategies concentrated on maximizing the consumption of resources while reducing the complexity of the calculations required to achieve this goal.
P1-15
Effects of Subject-Defined Categories on Discrimination Performance in Confidence Assessment Tasks
Glenn J. Browne, Shawn P. Curley, & George Benson (University of Maryland--Baltimore, University of Minnesota, & Rutgers University)
In this study, subjects created their own confidence assessment categories and located them at any point on the confidence scale. Using traditional quality measures, we tested judgmental performance by comparing subjects' categories against experimenter-defined categories with equal widths of .10. We also tested whether subjects located assessments more toward one end of the scale or the other. Further, we tested whether subjects could accurately determine the number of confidence categories into which they can discriminate. Subjects' judgmental accuracy is not generally good according to calibration measures; this study provides evidence for self-knowledge of discrimination ability.
P1-16
The Effects of Goal Setting and Performance Feedback on Automation Bias
Mark D. Burdick & Linda J. Skitka (Department of Psychology, University of Illinois at Chicago)
Can goal setting and performances feedback be used as an effective cognitive de-biasing technique to reduce the incidence of automation bias, i.e. the tendency to make errors of commission (inappropriately following automated directives) or of omission (failing to take action because an automated aid failed to alert the operator to the presence of an error)? One hundred eighty-one subjects received either accurate, inaccurate or no performance feedback. Results indicated that feedback significantly reduced commission errors but had no effect on omission errors. Implications are discussed.
P1-17
Effects of Structuring Aids, Knowledge, and Problem Characteristics on Solution Generation
Adam Butler & Lisa Scherer (University of Wisconsin & University of Nebraska at Omaha)
In this study, the manner in which objectives were presented was examined as a solution generation aid for two ill-structured problem scenarios. High knowledge individuals generated more solutions and more solutions that resolved the conflict for both problems compared to low knowledge individuals. When two objectives were presented in a decomposed form, participants generated more solutions than when the same two objectives were presented simultaneously. For a problem about compensation, participants presented with decomposed objectives also generated more resolving solutions than participants who were not presented with any objectives; however, the quality of solutions for a problem about sexual harassment was unaffected by the presence of structuring aids.
P1-18
The Role of Self-Efficacy on Information in a Negotiation
Matthew V. Champagne & Thomas Tabaczynski (Dept. of Cognitive Science, Renesselaer Polytechnic Institute)
Although it is clear that prior information plays a strong role in decisions made by parties during a negotiation, it is unclear what types of information result in certain outcomes and by what mechanism this information affects a decision. We propose that information affects the self-efficacy or confidence of the bargainer and this, in turn, determines the initial bid and eventual outcomes. Self-efficacy for a negotiation task was increased through the use of an "inside scoop." Results demonstrated that it was not the information per se, but rather the parties' confidence that accounted for the outcome of the negotiation.
P1-19
Visual Positioning and the Relationship to Consumer Decision Making
Bennett Cherry, Lisa D. Ordonez, Susan Heckler, & Barry Sheppard (Depts. of Management & Policy and Marketing, University of Arizona, and SHR Perceptual Management, Scottsdale, Arizona)
This study investigates how the presentation of an object influences the perception of the object (or the "Visual Position"), thereby influencing the purchase decision. Of particular interest are the visual components of a product's photograph (i.e. lighting, close-up, angle). Stimuli consisted of 18 photographs of the same object constructed from a 6 (lighting types) x 3 (close-up levels) design. Subjects rated the 18 stimuli on purchase likelihood, likableness, quality and several other attributes. Additionally, subjects rank-ordered all stimuli on the subjective dimension of quality. A perceptual map of the stimuli will be generated to assist in determining how the visual components (lighting type and close-up level) affect perceptions and purchase decisions.
P1-20
Comparison of Frames for Choosing a Social Welfare Function
Todd Davies (Koc University, Istanbul)
Harsanyi (1955) and Rawls (1959;1971) have both argued that resource allocation can be guided by considering what distribution rule potential recipients of resources would choose before they know who they are. Rawls argues that such a veiled participant (VP) would be more inclined toward the "maximin rule" (maximizing one's resources in the worst case) than would an impartial spectator (IS), while Harsanyi argues that the VP would maximize expected utility. In an experiment using measured fractional utilities for interpersonal comparison across consumer items, VP subjects (potential item-recipients) were less inclined toward the maximin rule than were IS subjects.
P1-21
The Effect of Question Format on Beliefs and AIDS Risks.
Wandi de Bruine & Baruch Fischhoff (Department of Social and Decision Sciences, Carnegie Mellon University)
Although an efficient means of data collection, closed-ended questions may overstate the knowledge of those respondents who are able to guess the correct answer by reading between the lines; they may understate the knowledge of respondents who cannot express their beliefs in the terms offered. Open-ended questions, on the other hand, also face several potential problems. This study examines the effect of different question formats on responses to knowledge questions about AIDS, for several different audiences (i.e., low-risk teens, low-risk adults, and high-risk teens). Effects of item structure were found to vary by audience. Recommendations for knowledge assessment are suggested.
P1-22
Oiled Seabirds, Ice Cream, and Applesauce: More Similar Than You'd Think
Shane Frederick & Baruch Fischhoff (Dept. of Social & Decision Sciences Carnegie Mellon University)
Contingent values are often insensitive to the amount of good(s) offered for hypothetical sale. This has typically been attributed to respondents' unfamiliarity with buying or "consuming" the exotic goods valued in CV studies (e.g., restoring x miles of oiled beaches). However, we found that respondents' willingness to pay was insensitive to stated quantity even for common consumer goods such as applesauce, ice cream, toilet paper, and canned tuna. This result casts doubt on the possibility of obtaining sensible values of public goods, using traditional CV methods.
P1-23
The Role of Rater Similarity in the Judgement of Wisdom
Barbara DeFilippo (Linn-Benton Community College)
Judgments of ambiguous traits, such as wisdom, appear to be a complex interaction of the traits of both the rater and target. Previous research has indicated that wisdom judgments are not based on values of a constant set of traits. Rather, the role of the target, and gender of both rater and target, influence which traits are related to wisdom ratings. In this research, the role of similarity on the judgment of wisdom is explored. Pilot data suggests that traits the rater possesses him or herself are more likely than those not possessed to be used in determinations of wisdom.
P1-24
Is Clinical Hypothesis Testing Influenced by the Framing of Hypothesis?
Michael E. Doherty & Kenneth M. Shemberg (Department of Psychology, Bowling Green State University)
Research shows that people are biased when testing hypotheses; the clinical literature suggests that clinicians are biased to see psychopathology. Clinicians (N=102) read a biographical essay. They judged the writer as psychologically healthy or not, psychologically healthy or unhealthy, or as experiencing psychopathology or not. The latter condition was expected to produce the most negative judgements. The opposite obtained. To determine if this was due to extreme term, "psychopathology," 29 clinicians assessed whether the writer was pathologically unhealthy or not. Their data were like the two less extreme conditions. These clinicians were conservative in judging the presence of psychopathology.
P1-25
Rethinking Overconfidence Yet Again
Gregory L. Brake, Michael E. Doherty, & Gernot D. Kleiter
(Bowling Green State University & University of Salzburg) Two experiments are presented. In the first, fifty subjects answered 150 general knowledge questions, plus 50 replicates from the original set. Although the effect is weakened, overconfidence was still found when a randomly sampled set of questions is used. Furthermore, there was only moderate reliability (r = .70) in subjective probability (SP) estimates. The second experiment employed a novel task. Twenty subjects who scored very high on a test of baseball knowledge were presented with rich but incomplete descriptions of 150 randomly sampled baseball games, asked to predict the winner of each game, and assign an SP thereto. As in the first experiment, a set of replicates was also presented to the subjects. Substantial underconfidence was found. There was high reliability (r = .82) in SP estimates. One explanation offered for the overconfidence effect typically found is alternative hypothesis neglect. That is, judges tend to only consider a single hypothesis and, as a result, generate an SP for that hypothesis that is much higher than they should be. We propose that the underconfidence found in the second study is a result of subjects attending to evidence for both hypotheses. In a rich environment like that used, subjects are given a great deal of information in favor of both hypotheses, resulting in lowered probability estimates (and, hence in this extreme case, underconfidence).
P1-26
Are Kids Smart Gamblers? A Developmental Time line to a Frontal Lobe Gambling Task
Michael Emerson & Jerry W. Rudy (Department of Psychology, University of Colorado at Boulder)
We investigated the differences in performance between adults and children on a gambling task. The gambling task simulates real-world decision making and has been used to differentiate frontal lobe patients with damage in the ventromedial prefrontal cortex from unimpaired people. We hypothesized that children would perform differently than adults, even when taking mathematical and monetary knowledge into account. The participants included 35 college students and 16 elementary students. Results indicated that our hypothesis was correct. However, the learning curve exhibited by the elementary students indicated that they may not have understood the gambling task. A modification to make the gambling task more concrete is suggested for further research.
P1-27
Regret and Disappointment in Decision Making
Leonie E. M. Gerritsen and Gideon B. Keren (Eindhoven University of Technology)
We have tested some predictions that follow from Loomes and Sugdens' regret and disappointment theory. Whereas they tested their theories in a gambling paradigm, we employed more realistic and common settings. The two major findings of our studies that are incongruent with the framework of Loomes and Sugden are: (i) experienced disappointment is strongly influenced by outcomes of non-chosen options; and (ii) regret also depends on previously set expectations, thus also probability. An alternative framework to account for these findings will be presented.
P1-28
Head or Heart Decisions: Which Do We Regret More?
Karen Steinberg (Department of Psychology, University of Pennsylvania)
Head-heart decision conflict occurs when people feel torn between two mutually exclusive options: the apparently rational (or "head") choice and the seemingly irrational (or "heart") choice. Subjects were asked about a head-heart decision conflict from their own lives. At the time of making the decision, those who had heeded their heads expected to regret the decision more in the short term and less in the long term; those who had followed their hearts anticipated the reverse. In actuality, both groups regretted their decisions the same amount, which was the lesser of the two amounts they had anticipated experiencing.
P1-29
"If Only" Thinking: Upward Counterfactuals, Regret, and Depression
Karen Steinberg (Department of Psychology, University of Pennsylvania)
Does "if only" thinking--the tendency to imagine better outcomes to negative events, or to make upward counterfactuals--predict depression? Subjects listed their thoughts about negative situations, six hypothetical and one from their own lives. Although both the number and proportion of upward counterfactuals correlated with a measure of regret-proneness that predicted depression, they did not correlate with a measure of depressive symptoms. The results suggest that the underlying link between regret and depression is not a tendency to imagine how things might have been better but to react to negative events in a pessimistic, self-blaming way.
P1-30
The Relationship Between Time Perspective, Sexual Behavior and Predictions of Contraceptive Use
Karen E. Jacowitz (Department of Psychology, UC Berkeley)
This study investigated whether individual differences in present and future time perspective were related to sexual and contraceptive choices. The theory of planned behavior was used to predict intentions to use four contraceptive methods and self-reports of actual contraceptive use. Results showed that scores on present but not future time perspective were related to some sexual behaviors. However, contrary to predictions, scores on the time perspective scales were not significantly related to the ability of the model to accurately predict intentions or behavior.
P1-31
Temporal Dimensions of Health Risks
Julie Goldberg & Alan J. Schwartz (Dept. of Psychology, University of California, Berkeley)
We investigated health risks relevant to college students. Risky behaviors varied not only in type (unprotected sex, drug use, or drinking and driving), but in two temporal aspects, past and future occurrences. In some cases, behaviors had been ongoing; in others, behaviors had never previously occurred. Similarly, behaviors would or would not be repeated in the future. A multidimensional scaling of judgments of risk similarity and evaluations of components of risk revealed that subjects were sensitive to temporal aspects, such as whether the risk resulted from one-time or repeated behaviors. We discuss implications for the study of health risks.
P1-32
Past and Future Possible Worlds: Evaluating Sequential Decisions
Alan D.J. Cooke & Alan J. Schwartz (UC Berkeley)
When selling a stock, your evaluation of the sale may depend on outcomes other than the one you actually received. You may also consider the values you might have obtained had you sold the stock earlier or later. In 4 experiments, subjects played a sequential investing game. They were shown values drawn from a random distribution and decided whether to accept the value or move to the next. Once they accepted a value, they were shown the next 0, 1, 2, or 3 future values. Past values had no effect on evaluations of elation or disappointment. Future values, however, had a strong negative effect on evaluations, and disappointing future information had a larger effect than did elating future information.
P1-33
Strategies and Cues When Making Group Judgements
Rebecca Henry, Deborah Ladd, Stefani L. Yorges, & Oriel J. Strickland (Purdue University, West Chester University, & California State)
The present study was an investigation of the strategies groups use to come to consensus as well as the cues individuals believe they rely on to evaluate the relative quality of members' inputs. Although outcome feedback did not have the anticipated effects on group accuracy, it did influence the strategies and cues individuals reported using when reaching consensus. Cash incentives also did not have the intended motivational impact, but did influence how group members felt about the quality of the social interaction of the group.
P1-34
Evidence of Gain/Loss Asymmetry in Employee Selection
Scott Highhouse & Michael Johnson (Departments of Psychology, Bowling Green State University and Indiana University-Purdue University Indianapolis (IUPUI))
We tested the propositions of Tversky and Kahneman's (1991) RDC model in the context of employee selection. In experiment 1, the undergraduates (N=82) were provided with information on two job candidates, along with (incomplete) information on a job incumbent. In experiment 2, undergraduates (N=74) received assessor training and then watched an example of either excellent or poor performance in one of two work samples used to predict employee success. In both experiments, candidates representing the status quo were preferred over candidates representing a loss, but candidates representing a gain were not preferred over candidates representing the status quo.
P1-35
Effects of Scarcity Information in Job Advertisements on Initial Job-Choice Decisions
Scott Highhouse, David J. Beadle, & Andrew Gallo (Indiana University-Purdue University Indianapolis (IUPUI))
Research has documented a tendency for people to assign higher value to objects that are scarce than to objects that are plentiful (e.g., Brock, 1968; Cialdini, 1988; Lynn, 1991). We examined the effects of scarcity information in the context of an entry-level job advertisement. Participants (N = 172) were presented with a job advertisement differing by vacancy availability (scarce vs. plentiful) and scarcity type (time vs. vacancy). Results showed that both time and vacancy scarcity had positive effects on pay inferences and company-image perceptions, but did not influence intentions to apply.
P1-36
Decision-Based Preference Formation Process
Steve Hoeffler & Dan Ariely (Fuqua School of Business, Duke University)
It is commonly accepted that preferences are constructed in order to resolve some of the ambiguity or conflict inherent in the decision task. The current work investigates the role that the choice process itself has on the formation and consolidation of these preferences. We hypothesize and support the idea, that repeated choices increase the confidence subjects have in their attribute importance ratings. That is, over time and repeated choice, the representation of the attributes become more precise and less malleable. We present three experiments that explore and support the theory of a decision-based preference formation process.
P1-37
Bayesian Reasoning in Ecological Contexts: The Impact of Information Representation for Physicians
Ulrich Hoffrage & Gerg Gigerenzer (Max Planck Institute for Psychological Research, Munich, Germany)
Is the mind, by design, predisposed against performing Bayesian inference? Previous research suggests the answer is yes. However, any claim against the existence of a particular cognitive algorithm, Bayesian or otherwise, is impossible to evaluate unless one specifies the information format in which the algorithm is designed to operate. We demonstrate that frequency formats entail Bayesian algorithms that are computationally simpler than those entailed by the probability formats previously used. We analyzed the cognitive algorithms of 48 physicians working on medical problems presented in a probability or a frequency format. As predicted, performance was much better in a frequency format.
P1-38
Missing Attribute Versus Missing Information: Further Evidence of Inferring Missing Information
Carolyn M. Jagacinski (Dept. of Psychological Sciences, Purdue University)
In this experiment, raters were asked to evaluate cars described in terms of ratings of their safety and the performance of their computers. The correlation between these sets of ratings was described as positive, negative or zero. Some cars were missing ratings and some did not have a computer. The evaluations of cars described only in terms of safety differed by correlation condition when the computer rating was missing but not when the car did not have a computer. The results suggest that raters inferred the computer rating when it was missing.
P1-39
The Relationship Between Choice and Confidence Response Times and Confidence Judgements
Jeremy D. Jokinen & Bruce W. Carlson (Psychology Department, Ohio University)
First, subjects predicted the winner of 40 pairs of football teams. 10 pairs of football teams were presented 10 times, 5 times, 1 time, and 10 pairs were not presented. Second, subjects were asked to again make predictions for the 40 pairs of football teams and then indicate their confidence. Time to make a prediction, time to select a confidence judgment, and confidence were measured. Results indicate a linear decrease in time to make a prediction and a linear increase in confidence for predictions subjects were exposed to 0 times, 1 time, 5 times, and 10 times.
P1-40
The Joint Effects of Negative Affect and Task Difficulty in Multi Attribute Choice
Kathryn Kadous & Dan N. Stone (University of Waterloo & University of Illinois at Urbana-Champaign)
Can emotions improve decision performance in some tasks but impair it in others? We theorize that increasing negative affect leads decision makers to process information faster but less carefully. We find support for this theory in an experiment in which 255 participants chose colleges in conditions created by crossing two levels of negative affect (high, low) with two levels of choice difficulty (hard, easy). Our results include: (1) participants under conditions of high negative affect processed information faster but less carefully, and (2) these changes in decision strategies increased choice accuracy in easy tasks but decreased it in hard tasks.
P1-41
Probability Judgment in Three-category Classification Learning
Derek J. Koehler (University of Waterloo)
People generally give subadditive probability judgments when they assess each in a set of three or more exclusive hypotheses. The degree of subadditivity in such judgments is determined in large part by the evidence upon which the judgments are based, but the characteristics of the evidence that influence subadditivity have yet to be fully specified. This issue is addressed through the use of a three-category classification learning task, in which the relationship between the evidence and the hypotheses under consideration can be controlled experimentally. Two potential evidential influences on subadditivity, cue conflict and cue frequency, are distinguished and tested.
P1-42
An Alternative Decision Making Methodology
Creig Kronstedt (Cardinal Stritch College)
Establishing accurate behavioral decision making norms requires the elimination of personal biases. The present research elicited responses to a set of ethical behavior decisions from adult business students at Cardinal Stritch College, Milwaukee, WI. It then asked participants to evaluate the ethicality of business associates, business competitors, and finally of themselves. Regression analysis of the decisions accounted for seventy percent of the variability of perceptions of the ethicality of others, but was only able to account for fifty percent of the variability of participants' perceived ethicality. The methodology proposed is offered to enhance validity of decision making research.
P1-43
The Impact of Damage Caps in Civil Litigation
Greg Pogarsky & Linda Babcock (Carnegie Mellon University-The Heinz School of Public Policy)
A major issue in the current debate over civil justice system reforms is whether to limit the amount of damages a plaintiff can receive. The economic model predicts such "damage caps" will reduce the likely disparity between parties' expectations of the trial outcome and, hence, increase the likelihood of a pretrial settlement. Here we present two alternative psychological theories, each yielding predictions quite distinct from those of the economic model. This paper will report the results of negotiation experiments designed to test the impact of "damage caps" on litigants' pretrial negotiating behavior.
P1-44
The Effect of a Survival Versus a Success Focus on Decision Making Under Risk
Terry L. Boles & Rashmi Gupta (Department of Management and Organizations, University of Iowa)
This study examines risk preferences in investment decisions when there are two salient reference points; (1) an aspiration level ( that a hypothetical organization had to meet to insure continued success) and, (2) a survival level (that the organization had to stay above to remain solvent). Varied were the current wealth of the organization ( 4 levels), investment amount (3 levels), and amount of return on investment (2 levels). Respondents indicated how high the probability of success would have to be before they would invest. Across all investment and return amounts, those whose current wealth was well above both reference points were willing to assume the most risk, however those at the other end of the distribution (who were near bankruptcy) were equally risk seeking up to the point where investment amount would make them insolvent, then they switched to risk averse behavior. We discuss the conditions under which foci other than current wealth affect decision making under risk.
P1-45
Attribute Frame Effects: The Impact of Direct Experience
Kathryn A. Braun & Gary J. Gaeth (University of Iowa)
This study investigates differences found in past framing research. Ss evaluated two types of frozen yogurt- one low-fat, the other very low in fat. The yogurts were either both positively framed- 90% fat- free, 99% fat-free or negatively framed- 10% fat/ 1% fat. Half the Ss tasted the yogurts, the other half did not. Both groups evaluated and made a choice between the two yogurts. Consistent with prior framing research, yogurts were rated more favorably under the positive frame, less favorably under the negative frame. We found the negative frame more influential in choice in the healthier yogurt, but some level of involvement was necessary. The taste experience diminished those effects.
P1-46
Complexity, Context, and Constraints in Human Decision Making
Ray W. Cooksey (Department of Marketing & Management, University of New England, Australia)
This paper develops a systems-oriented complexity perspective on human judgment and decision making. Thinking in domains such as general systems theory, chaos/complexity theory, fuzzy set theory, and biological/evolutionary psychology have advanced to the state where they can be brought to bear on an understanding of human decision making as it unfolds in real time under constraint. Adopting this perspective leads to a view of human decision making which is constraints-bounded, highly sensitive to the initial contextual conditions which predominate prior to the decision act, and subject to nonlinear dynamic processes.
P1-47
The Directional Relationship Between Expectation and Behavior Rachel Croson & Mark Miller (The Wharton School,
University of Pennsylvania and Social & Decision Sciences, Carnegie Mellon University)
Experimental economists and psychologists have noticed a relation between the expectation that an individual has about their counterpart's behavior in a game and the individual's own behavior. Specifically in the Prisoner's Dilemma game, those expecting cooperation are more likely to cooperate themselves, and those expecting defection are more likely to defect. Does expectation cause behavior, or does behavior cause expectation? We ran an experimental game to test the direction of the relation.
P1-48
Eliciting Importance of Attributes, Still an Easy Task?
Orfelio Leon
Should SMART be dead--as Edwards and Barron affirmed in 1994? This poster is a comparative study of the methods SMART and SMARTS, in relation to eliciting weights. To these two techniques has been added a graphical version of PA (GRAPA--GRAphical Point Allocation). ("Grapa" means "staple" in Spanish.) This technique consists of asking the decision-maker to divide a small set of counters (5 x number of attributes) among as many columns as there are attributes of the problem.
P1-49
Dependent vs. Independent Confidence: Implications for (Non)Additivity of Subjective Probability of Two Hypotheses
Craig R.M. McKenzie (Department of Psychology, UC San Diego)
This research distinguishes between dependent and independent confidence in two mutually exclusive and exhaustive hypotheses. With dependent confidence, a change in confidence in one hypothesis results in a complementary change in the other; thus, one would expect subjective probabilities in the two hypotheses to be (roughly) additive, or sum to 100%. With independent confidence, however, a change in confidence in one hypothesis is not necessarily accompanied by a complementary change in the other. Confidence can change in one hypothesis, but not the other. In fact, confidence can increase (or decrease) in both hypotheses simultaneously.
P1-50
Framing Effects and Output Interference in an Audit Planning Context
Karla M. Johnstone, Jean C. Bedard, & Stanley F. Briggs (Department of Accounting, University of Connecticut and Northeastern University)
This study investigates the independent and interactive impact of framing effects and output interference in an audit planning context. An experiment involving a financial statement analysis task and a complex contractual revenue recognition solution task for an audit client with significant government contracts provides data to test predicted differences in auditor performance due to framing effects and output interference. Unlike prior studies that have examined framing effects and output interference in isolation, this study explores the similarities and differences of these two judgment biases and provides evidence regarding their interactive effect. The results indicate an interactive effect of output interference and framing effects, but output interference appears to outweigh the impact of framing effects.
Poster Session #2: Sunday Nov. 3, 5:30 - 7:00 pm
P2-1
Desire Not to Waste Versus the Desire for New Things
Hal R. Arkes & Laura Hutzel (Department of Psychology, Ohio University)
The desire not to waste and the desire for new things are manifested in contradictory behaviors--pro- environmental in the former case and anti-environmental in the latter. Five pairs of questionnaires solicited subjects' willingness to build a new facility versus renovate an old one. We show that to the extent nature is cued, the desire not to waste is dominant, thus resulting in renovation of the old structure. Otherwise the desire for new things in dominant. Additional data suggest this latter desire may be a result of the application of Akerlof's Lemon Principle: used items are potentially defective.
P2-2
Base-rate Fallacy: Heuristics, Frequent Phrasing and Pragmatic Factors
Laura Macchi (Istituto de Psicologia, Universita' Di Milano)
The present study reconsidered the frequentistic approach to the base-rate fallacy (Tversky & Kahneman, 1973, 1980) given by Gigerenzer (1991, 1995). We showed that other factors, apart from the frequentist phrasing itself, have been introduced with the frequentist reformulation, and that they are responsible of the disappearance of the base-rate fallacy in some classical problems(Cosmides & Tooby,1996). Independently on the presence or absence of at least some heuristics and frequentist phrasing, base-rate neglect can occur or not occur, depending on some pragmatic factors, such as particular text formulations which induce or eliminate confusion between likelihood and the posterior probability.
P2-3
Modelling Consumer Response to Ambiguity in Retail Price Advertisement
Sanjay Dhar, Claudia Gonzalez-Vallejo, & Dilip Soman (College of Business & Administration, University of Colorado at Boulder)
Advertisements for retail promotions often imprecisely specify the level and chances of getting the advertised discount (a tensile discount). In this paper we identify and test factors that consumers use in making inferences about the probability of getting a discount. We develop two models that capture the consumer's response to tensileness and compare them experimentally. Finally, we combine the results from the inferred probability framework and the contingent weighting model (Sattath, et al, 1988) to show that for every retail store, there exists a threshold discount level above which tensile claims will outperform precise claims and below which precise claims will outperform tensile claims.
P2-4
Changing Preferences: A New Probabilistic, Context Sensitive, and Intransitive Model of Choice Behavior
Claudia Gonzalez-Vallejo (Center for Policy Research, SUNY Albany)
The fluctuating nature of our preferences is modeled via the proportional difference (PD) model. PD assumes that decision makers trade attributes in proportions and that this trading is a random process, sometimes leading to intransitive preferences. The probability of selecting one alternative over another is a function of the proportional difference, d, and a decision threshold, which allows for differential weighting of the stimuli's attributes. The model produces intransitive choices, predicts interesting context effects, and can deal with vague information in its multiattribute form. The model was fitted to data in four experiments, and its behavior compares to predictions of contingent weighting, the contrast weighting model (Mellers, et al, 1992) and an expected value threshold model. The simple two-parameter PD outperformed all two-parameters models.
P2-5
Judgments versus Decisions Based on Frequency Information
Craig McKenzie & Jason S. Mills (Department of Psychology, University of California - San Diego)
This research distinguishes between dependent and independent confidence in two mutually exclusive and exhaustive hypotheses. With dependent confidence, a change in confidence in one hypothesis results in a complementary change in the order; thus, one would expect subjective probabilities in the two hypotheses to be (roughly) additive, or sum to 1, or 100%. With independent confidence, however, a change in confidence in one hypothesis is not necessarily accompanied by a complimentary change in the other. Confidence can change in one hypothesis, but not the other. In fast, confidence can increase (or decrease) in both hypotheses simultaneously. Thus, one would not necessarily expect additivity to hold. In order to test for the distinction between dependent and independent confidence, subjects learned to diagnose two illnesses. By viewing patient profiles listing the presence or absence of four symptoms, one group of subjects learned to distinguish between the illnesses (the contrastive learners, or CL subjects), and one group learned about the illnesses separately (the noncontrastive learners, or NCL subjects). CL was predicted to lead to dependent confidence, and NCL to independent confidence.
P2-6
Rent-to-own Revisited
Andrew Parker & Brian J. Zikmund-Fisher (Dept. of Social & Decision Sciences, Carnegie Mellon University)
Over the past decade, more and more people have turned to rent-to-own as an important purchase option. Controversy over such contracts centers on the extreme implied interest rates and the fact that usage is limited almost exclusively to the poor. Discounting and other traditional economic explanations do not completely explain this phenomenon. Following up on results presented last year, we demonstrate the importance of various issues on the attractiveness of renting-to-own to the poor, including income variability, difficulty in saving, self-control, feelings of debt, and pre-commitment mechanisms.
P2-7
Using Judgments to Understand Decoy Effects in Choice
Jonathan C. Pettibone & Doug Wedell (Department of Psychology, University Of South Carolina)
Students were presented choice triads, with alternatives described along two dimensions. In Experiment 1, the decoy alternative in each set was dominated by only one of the two other alternatives. In Experiment 2, the decoy was dominated by both alternatives. Within different blocks of trials, participants rated
P2-8
Effects of Information Presentation Constraints on Choice
Stuart M. Senter & Doug Wedell (Department of Psychology, University of South Carolina)
Students chose from sets of apartments described along different dimensions. Information was presented in either an alternativewise, dimensionwise, or completely unconstrained fashion. Apartment sets were designed to favor different alternatives for dimensionwise versus alternativewise strategies. The study sought to validate distinctions between these two types of strategies through examining both decision behavior and looking times. Under the alternativewise processing constraint, decision makers spent much more time looking at information and were more likely to choose inappropriate alternatives. These results provided support for the assertion that alternativewise processing is inefficient and suboptimal in a complex decision making task.
P2-9
The Relation Between Instrumental and Emotional Evaluations in Decision Making
Hans-Ruediger Pfister & Gisela Boehm (Technische Universitaet Berlin & University Bremen, Germany)
Three experiments are presented which investigate the role of emotions in decision making. It is proposed that choices are jointly determined by what we call instrumental and emotional evaluations. First, it was confirmed that both components contribute independently to the prediction of preferences. Second, it was examined if instrumentalities can be reduced to emotions, and vice versa; however, the relation seems to be highly domain dependent. Third, factors which determine the relative impact of both components were identified. Situational context such as public versus private decisions, and task characteristics such as response mode systematically change the decision maker's evaluative focus.
P2-10
Signal Detection Theory and Decision Making in Forensic Science
Victoria Phillips & Michael J. Saks (Psychology Dept., The University of Iowa)
Given the "fuzzy" and often incomplete nature of forensic evidence, forensic scientists are often forced to render judgments under ambiguous circumstances. Signal Detection Theory (SDT), a direct offspring of statistical decision theory, is well-suited for analysis of such ambiguous decision scenarios. According to SDT, the correctness of a decision is comprised of two distinct components: (1) the decision maker's diagnostic ability and (2) psychological factors involved in establishing a decision cutoff point. We argue that SDT analysis of forensic decisions could greatly advance the understanding of forensic examinations by dissociating and quantifying decisional accuracy from inherent decisional biases.
P2-11
Social Context and Individual Differences in Decision Making
Victoria Phillips & Irwin P. Levin (Psychology Department, The University of Iowa)
The social context of human judgment often influences decision processes and outcome(s). The current study addresses potential mediators of social influence. Participant differences in Self-Monitoring (SM) were examined as indicants of differential susceptibility to social influence. Mediating effects of prior knowledge and information order were also examined across two different judgment tasks: 1) personal preference and 2) recommendation for another. Results suggest that differences in self-monitoring correspond to differences in sensitivity to social information. Findings highlight the importance of considering both social context and individual differences in decision models.
P2-12
Measuring Personal Values in Breast Cancer Treatment Decisions: Does It Matter How You Ask The Question?
Penny F. Pierce (School of Nursing, The University of Michigan)
This poster describes a pilot study designed to determine if the format and structure of questions concerning personal values influences the ratings of women hypothetically selecting a breast cancer treatment. Women seeking mammography were presented one of three randomly selected scaling techniques to assign values to various dimensions of treatment. Results provide insights into the ways in which women's choices are influenced by their personal values as well as the influence of the structure of the measurement on their ease or difficulty resolving conflicts among competing values. Suggestions are provided concerning the development of a clinically useful and reliable measure.
P2-13
Student Nurses' Perspective-taking Judgments of Medication Acceptance
Celia E. Wills & Colleen F. Moore (College of Nursing, Michigan State University Department of Psychology, University of Wisconsin- Madison)
Student nurses (n=29) made judgements of the likelihood of medication acceptance from the perspectives of three hypothetical others who placed the most importance on one of three types of cues: level of trust in the medication prescriber, severity of a hypothetical mental health condition being experienced, and the potential side effects of the medication. The perspective-taking judgements showed the largest effect for the cue that was said to be most important for a given perspective. However, the nursing student perspective-taking judgements showed systematically more extreme effects of the most important cue for a given perspective, than actually occurred for a separate sample of undergraduate students (n=155) who made the same judgements based on their own opinions and who rated the three cues as to their relative importance in their judgements. Results will also be evaluated in relation to recently-proposed models of judgement processes for partially-described information (Wills & Moore, 1996). The findings have implications for how health care providers infer people's preferences and patterns of information use in collaborative health-related decision making contexts.
P2-14
Incentives for Organ Donation: Neither Unethical Nor Discouraging
J.D. Jasper, Jack Hershey, Carol Nickerson, & David A. Asch (Philadelphia Veterans Affairs Medical Center and University of Pennsylvania)
Our purpose was to investigate reactions toward various incentive policies designed to encourage organ donation. 300 prospective jurors completed one of four surveys, each providing a different context in which one might consent to donation. Each survey described the current system of unrewarded donation and 6 to 8 potential incentive policies. Contrary to previous claims, Ss did not consider most incentive policies to be unethical, not did they indicate that they would be less likely to donate under such policies. Nonmonetary incentives and those designed to ease the burden of the family were most popular.
P2-15
The Influence of Information on Belief Processing by Arthritis Patients
Noel Wilkin & Glenn Browne (Department of Information Systems, University of Maryland, Baltimore)
Drug consumption behavior influences the effectiveness, cost, and outcomes of drug therapy. Beliefs significantly influence people's health behaviors. Information is the principal factor used by health care professionals to influence those beliefs. This study used the model of belief processing proposed by Smith, Benson, and Curley (1991) to examine the impact of belief information in a health care context. The researchers expect the provision of information to influence the arguments used by arthritis patients in forming a belief about the helpfulness of a medication. An understanding of belief formation will ultimately contribute to developing interventions that effectively influence health behaviors.
P2-16
Does Diagnosis Make You Sick? Disease Knowledge and Health Self-Assessment
Terry Connolly & Adrianne Casebeer (University of Arizona & Colorado State)
Decision regret is generally thought to involve remorse or self-blame, as when active choice leads to self-recrimination as well as poor outcomes. In five experiments we compared participants who experienced identical gains or losses as a result of either their own choices or of an external, arbitrary process. Final outcome evaluations were heavily path-dependent, but decision agency showed no effect. Regret thus appears not as a special emotional component involving self-blame but simply as a label for the disutility associated with a decline in fortunes.
P2-17
Reasoning Patterns in Tactical Problem Solving
Jon J. Fallesen & Julia Pounds (ARI Research Unit, Leavenworth and U.S. Army Institute)
Tactical problem solving in uncertain and dynamic environments generally encompasses a variety of risks and high stake consequences. Prior analyses demonstrated that planning under these circumstances may not be sufficiently characterized by a particular problem solving or decision making approach. Thus, the current line of research reframes tactical problem solving as a reasoning process. Verbal protocols collected during interviews with 78 military leaders were analyzed. Findings illustrate how tactical planners reason about problems that include missing information.
P2-18
Self Reported Performance Satisfaction Levels: A Question of Deception
Adrian Rantilla, Bradley D. Crouch, & James H. Davis (University of Illinois at Urbana-Champaign)
In many studies, researchers arbitrarily select deceptive feedback values. In this research, normative values were obtained from 412 subjects who indicated percentile judgments they considered to denote poor, average and excellent performances. In study 1, 125 subjects first gave the excellent score (High Anchor Condition). In study 2, 287 subjects began with the poor score (Low Anchor Condition). Overall, the mean poor, average and excellent scores were 50.58 (SD=19.83), 69.06 (SD=13.29), and 86.07 (SD=9.36), respectively. High and Low Anchor Condition means differed significantly on both the poor and average scores. In all cases, High Anchor means were greater than Low Anchor.
P2-19
Choosing Public Goods: Exploring the School Choice Decision Process
Kenneth Rona & Mark Schneider (Fuqua School of Business, Duke University and SUNY Stony Brook)
The delivery of services by local governments involves a complex relationship between the institutions that supply them and the citizens who use them. Many reform proposals argue that governments should imitate private markets by increasing the number of suppliers of public goods and then "empowering" citizens to shop among the expanded choice set. This idea has been proposed in a number of policy domains, especially schools. This paper examines two factors, that may affect parents' school choice process, search costs and the number of alternatives in the choice set. We conducted a process tracing experiment that presented parents with hypothetical schools. Our results indicate that, when faced with small search costs, parents shift toward selective decision strategies and actual choice is less congruent with predictions of a weighted additive model (WADD). The number of alternatives manipulation had no effect on selectively but did decrease choice congruence with the WADD model. Finally, the implications for school choice proposals are developed and a more detailed research program is proposed.
P2-20
Does Evidence Presentation Affect Judgement? An Experimental Evaluation of Five Display Formats
Alan Sanfey & Reid Hastie (Cognitive Psychology, University Of Colorado)
Many everyday judgment tasks can be conceptualized as reasoning from items of information to estimate a magnitude, quantity, or condition. Experimental participants made judgments of marathon finishing times based on information about the runners' ability, training, and motivation. The information was displayed in one of five formats: a table of numbers, two types of bar graphs, a brief text, and a longer biographical story. The various information displays yielded different patterns of cue utilization and accuracy. In accordance with the Explanation-Based Decision Model, the text and story displays produced the most accurate judgments of marathon finishing times.
P2-21
Mood As Input To Counterfactual Thinking
Lawrence J. Sanna & Kandi Jo Turley (Department of Psychology, Washington State University)
Past research has shown that downward counterfactuals (alternative post-outcome thoughts about "what might have been" that are worse than actuality) produce good mood and upward counterfactuals (alternative post-outcome thoughts about "what might have been" that are better than actuality) produce bad mood. In the current set of studies, borrowing from mood-as-input reasoning, we demonstrate that the opposite relationship is also true: good mood produces downward counterfactuals, and bad mood produces upward counterfactuals. Implications for the role of counterfactuals in decision making, and the potential bi-directional interrelationships between counterfactuals and mood, are discussed.
P2-22
Layoff Implementation Decision Making
Donald H. Schepers & Stephen Gilliland (Department of Management and Policy, University of Arizona)
Current research has focused on the impact of managerial actions during a layoff on the attitudes and behaviors of victims and survivors. This research examines managerial decision-making during layoff implementation. Subjects (n=113) were asked to read a scenario that varied size and level of responsibility for the layoff. Results showed that these variables affected implementation decisions along the dimensions of method of informing employees, amount of information given to employees, and amount of personal involvement of managers in the layoff process.
P2-23
Temporal Changes in Specific Affective States: Implications for Decision Making
Lisa L. Scherer, R. Jason Weiss, & Judy A. Condon (Department of Psychology, University of Nebraska at Omaha)
Empirical investigations have produced conflicting results regarding global affective influences (positive versus negative) on decision making. These inconsistencies may be explained by important yet ignored differences in how specific types of negative or positive affect are influenced and change over time. Sixty-eight undergraduates responded to four affect scales positioned at the beginning, middle or end of a 45-minute questionnaire in a pleasant or unpleasant room, varying in odor, lighting, appearance, and comfort. As predicted, ANOVAs revealed statistically different patterns of change over time for positive affect type (happy versus lively) and negative affect type (bored versus irritated).
P2-24
Effect of Solution Generation on Cognitive and Affective Reactions to Problems
Lisa L. Scherer, R. Jason Weiss, Roni Reiter-Palmon, & Deborah F. Goodman (Department of Psychology, University of Nebraska at Omaha)
Research has argued that the quantity and quality of solutions generated to ill-structured problems may be influenced by an individual's subjective reactions to the problem characteristics and the task demands operating on them. In this study, 240 participants reported their reactions to one of 12 ill-structured problems by responding to five affective and five cognitive scales both before and after generating solutions to the problem. Results showed that participant reactions to a problem changed after generating solutions for it. Specifically, personal involvement, negative arousal, fear, and perceived problem realism decreased, whereas elation and boredom increased following solution generation.
P2-25
Dividing the Loot and Claiming Debts Under Unequal Endowment
Gwen R. Grams, Ching-Fan Sheu, & David A. Kuhn (Department of Psychology, DePaul University)
We examine how people assign credit or blame for success or failure of group tasks. Subjects participate in die-rolling teams with anonymous partners. Team members receive initial endowments, and wager from $00 to the amounts of endowment. Each member rolls a die. The team with the highest total roll wins their total bet. Subjects assign credit/blame for the outcome to self and the partners. Results show subjects tend to assign more credit to themselves for winning and less blame for losing. The share of credit is influenced by die roll and endowment, and endowment interacts with the amount bet.
P2-26
Monetarily Compensating Victims of a Disastrous Event
Natasha Stein, Ralph Erber, Ching-Fan Sheu (DePaul University)
We study how people assign monetary compensation to victims of a disastrous event. A variety of scenarios were created to manipulate the following factors: the ages of the victims, the degree of innocence, and whether the presence of the victims in the situation is considered normal or not. The results show that innocent victims were assigned the most monetary compensation. The relationship between the ages of the victim and the monetary compensation was not linear. The youngest and the middle age tend to elicit more monetary compensation than do teenage victims and the elderly. The situation factor seems not to influence the assignment of monetary compensation.
P2-27
Multidimensional Inputs and Reward Allocation: Effects of Team-Building Instructions Studied With a Subtractive Model
Ramadhar Singh (Department of Social Work and Psychology, The National University of Singapore)
Instructions to "divide fairly" and "minimize conflict" in reward allocation have been believed to invoke use of the equity and equality rules, respectively. Instead I hypothesized that the allocation rule is subtraction and that the team-building instructions can engender response distortions on the allocation scale, cognitive distortions of inputs given, or both. Participants divided monetary reward between two persons of several groups "fairly" and in a way that can minimize group conflict. There were two inputs about each person--effort and performance (Experiment 1) or performance over two years (Experiments 2-4). As hypothesized, data of most of the individual participants conformed to the subtractive model. Patterns in the Instructions x Input levels effects across the raw and monotonically rescaled data further showed that the response distortions, input distortions, and response-input distortions are the most, moderate, and least frequent responses to the team-building instructions, respectively.
P2-28
The Distance Exclusion Model of Quantitative Group Decision Making
Paul Zarnoth & James H. Davis (University of Illinois at Urbana-Champaign)
Although several models of group choice among discrete alternatives have been introduced, there has been considerably less effort dedicated to understanding and predicting the judgments of groups from a continuum of options. In this paper, one model of quantitative group decision making, the Distance Exclusion Model (DEM) is proposed and tested in three studies. DEM assumes that if each group member has one equally-weighted vote, one optimum preference, and a preference curve which declines monotonically on each side of the optimum point, then the median individual judgment should be chosen by the group. In addition, models of group judgment should more precisely predict actual judgments if they consider the "functional" group size rather than the actual number of participants.
P2-29
The Effect of Depression on Overconfidence: Re-examining the Depressive Realism Hypothesis
Eric Stone & Carrie L. Dodrill (Wake Forest University & University of Houston)
The "depressive realism" hypothesis suggests that depressed people have more accurate perceptions about themselves and their environment; however, the empirical evidence regarding this hypothesis is equivocal. The present experiment examined depressive realism in the context of general knowledge (almanac) questions, and controlled for the potential confounding relationship between depression and anxiety. The primary finding was that, for non-anxious participants, depressives gave lower item-by-item and aggregate confidence judgments than did non-depressed participants. However, some evidence suggested that depressed participants also answered fewer items correctly, implying equivalent levels of overconfidence for non-depressed
and depressed participants.
P2-30
The Sunk-Cost Effect: The Importance of Context
Elmer Anita Thames (Psychology Department, John Carroll University)
The sunk-cost effect is characterized by an increased tendency to reinvest in a venture following an initial, irretrievable (or sunk) investment (Thaler, 1980). Study One examined the impact of two factors on the sunk-cost effect: the magnitude of the reinvestment cost and mental accounting. Subjects were asked to consider a scenario in which the object of initial investment was a ticket to a concert. As predicted, subjects were more likely to reinvest when the cost of reinvesting was low and when the reinvestment cost was applied to a different mental account from that of the initial investment. Study Two explored the possibility that the endowment effect might have influenced the results of Study One. In Study Two, subjects considered a scenario involving a weekend holiday at a hotel. The data suggested that endowment rather than mental accounting might account for the results of Study One. However, a contextual analysis is offered which might prove useful in understanding the relative influence of factors such as mental accounting, endowment, and
reinvestment costs on the sunk-cost effect.
P2-31
False Memories, Memory Judgments and Time: A Fuzzy-trace Theory Analysis
Allison L. Titcomb & Valerie F. Reyna (University of Arizona)
This study tested claims that false memories would not be observed for truth judgments and that gist and verbatim memories are assumed to behave similarly. A slide sequence was observed and recognition judgments, made immediately or after one week, showed differences in gist and verbatim memories as well as differences in truth and verbatim judgments. Truth judgments elicited larger false memory effects, although these effects varied over time for original versus false items and for different types of misinformation. Implications from fuzzy-trace theory include the need for early testing to ward off effects of forgetting on memory judgments.
P2-32
Expertise and Belief in Recovered Memories: A Decision- Making Analysis
Lori R. Van Wallendael & Karmen Mills (Dept. of Psychology, UNC Charlotte)
Randomly chosen adults, undergraduate and graduate college students, and psychology Ph.D.s completed a survey regarding their degree of belief in three recovered memory scenarios. All groups were responsive to the base rates of the incidents "remembered," with the greatest belief expressed in memories of being lost in a shopping mall. Groups were differentially responsive to a manipulation of recovery method: Experts expressed the greatest belief in memories recovered spontaneously upon revisiting the scene of the event, whereas groups with less psychology knowledge expressed the greatest belief in memories recovered under hypnosis.
P2-33
Psychological Factors Affecting Pain During Cold Pressor Experiences
Carol Varey (School of Cognitive & Computing Sciences, University of Sussex)
The effects of mood and social/counterfactual comparison on the experience of cold pressor pain were investigated in three experiments. A mood manipulation resulted in significant effects on pain rating during the experience and on retrospective evaluations of the total pain of the experience, but had no effect on minimum prices required to repeat the experience. Pain ratings, but not prices, were affected by beliefs about other temperature conditions in the experiment. Contrary to expectations, pain ratings were unaffected by beliefs about other durations - but prices were lower for 'shorter' experiences. Pain ratings and minimum prices appeared to be dissociated.
P2-34
Attentional Factors in Shoot/Don't Decision Making
David A. Washburn & Harold Greene (Center of Excellence for Research on Training, Morris Brown College)
Variables like visual search set-size, target eccentricity, distractor similarity are frequently examined in laboratory studies of attention and decision making. It is more difficult to manipulate important psychological variables in shoot/don't shoot scenarios, which are commonly filmed for training, not research, purposes. We tested volunteers with a series of scenarios in a firearms training simulator. Overt responses and eye movements (monitored with a head-mounted eye tracker) to the live-action stimuli were analyzed as a function of set-size and eccentricity. We will discuss these findings and the differences observed between simulator and traditional laboratory data.
P2-35
A New Methodology for Naturalistic Decision Making Research: Investigating Fire Officers' Decisions
Jim McLennan, Mary M. Omodei, & Alexander Wearing (Swimburne University, La Trobe University, & University of Melborne, Australia)
Researching naturalistic decision making processes has proved extraordinarily difficult, especially when the decisions are time pressured and stakes are high. Using a light-weight head-mounted video camera to capture a decision maker's own-point-of-view record of task activity, followed by video-assisted recall, enables real world decision processes to be investigated. Decision making by fire officers in command during large-scale simulation exercises were studies using the methodology. A system of coding decision processes was developed. Kelain's Recognition-Primed Decision (RPD) Model received support. However, other processes, such as means-ends analyses, were found to be important.
P2-36
The Heat of the Moment: Outcomes, Arousal and Risky Decisions
Ned Welch, George Loewenstein, & Stephanie Byram (Department of Social and Decision Sciences, Carnegie Mellon University)
Several researchers have proposed that emotions play an important role in decision making. Damasio's "somatic marker hypothesis," for example, posits that people use emotions as rough guides to the valence of alternatives. Little research, however, has examined the emotions evoked by different decision processes, such as the evaluation of outcomes, choosing between alternatives, or waiting for uncertainty resolution. We took physiological measures -- skin conductance and heart rate --during a series of decision making tasks involving risk. We examine changes in these measures associated with different phases of decision making, and test whether these measures predict subjects' choices.
P2-37
The Alternative-Outcomes Effect
Paul D. Windschitl and Gary L. Wells (Department of Psychology, Iowa State University)
People's beliefs that a focal outcome would occur were affected by manipulations to the distribution of alternative outcomes even though the summed probability of the alternative outcomes was held constant. Dividing the alternative outcomes into several small probabilities rather than a few larger probabilities increased perceived certainty that the focal outcomes would occur. The pattern of data across problem variations suggests that the effect is not due to changes in the number of alternatives per se, but rather to the presence or absence of a significant rival alternative. Discussion includes speculation on why this effect was previously undiscovered.
P2-38
The Effects of Differential Expected Value and Frame Variations Upon Gamble Preferences
Dave Wiseman (Department of Psychology, The University of Iowa)
Explicit assessments of the extent to which the expected value of a gamble is a moderating variable in the relationships between (1) employed valence description ("frame") of gamble outcomes and (2) gamble preference frequency are rare. Accordingly, the current study utilized five-outcome gamble and determined relationships among the above variables. Little relationship between frame and underlying risk attitude (as derived from paired preference data) was found for gambles with smaller expected values. In contrast, frame variation influenced preferences at larger expected values; risk averse preferences were exhibited more frequently when possible outcomes were described as gains than losses.
P2-39
Framing Effects Extended: The Case of Multiple Outcome Gambles
Dave Wiseman (Department of Psychology, The University of Iowa)
Past work assessing the influence of variations in the valence of presented gamble outcome descriptions (frame variations) upon choice has (1) used gambles which have, at most, two possible outcomes, and (2) failed to explicitly conceptualize risk magnitude as being a function of the variance and distributional inequality associated with employed gambles. This study used five-outcome gambles which varied in risk in accordance with the above construct conceptualizations. It was found that frame variations influenced gamble preferences across multiple (1) experimental designs, (2) decision contexts (financial/medical decisions), and (3) risk magnitudes. Subjects were less risk averse under negative frames that positive frames.
P2-40
Valence Effects as a Diagnostic Tool for Interpretation of Decision Task
Kimihiko Yamagishi (Faculty of Business & Environment, Shukutoku University)
Judgments of "How much better is a preferred option?" and "How much worse is a less preferred option?" may differ in their magnitudes. Such discrepancies are referred to as "valence effects." Previously, systematic positive valence effects ("Better" exceeding "Worse") were observed in the domain of gains, and negative valence effects ("Worse" exceeding "Better") in the domain of losses (Yamagishi & Miyamoto, 1996; Yamagishi, 1996). The current study extended this line of research by using valence effects as a diagnostic tool to assess decision maker's interpretation of choice tasks. Choice under "framing effect" (Tversky & Kahneman, 1986) was investigated. Framing effect ceased to occur when decision makers maintained consistent task interpretations (assessed from valence effects) as pertaining to gains or losses.
P2-41
Asymmetric Information Weighting in Belief Updating and the Effect on Mid-Sequence Evaluation
Gal Zauberman & Dan Ariely (The Fuqua School of Business, Duke University)
Despite the extensive research on belief updating, little is known about the way beliefs are updated while integrating continuous sequences of experiences. This study examines this issue by asking subjects to provide probability estimates after each new information item. Consistent with the conformation bias literature, we propose that new incongruent information will be underweighted following the formation of a hypothesis. Yet we also propose that this discounting will not occur in the initial stages of the process. A Bayesian approach was used to demonstrate the asymmetric utilization of the new information
before and after the hypothesis formation.
P2-42
Temporal Insensitivity in Judgments of Risk and Probability
David Fetherstonhaugh (Department of Psychology, Stanford University)
Several studies investigated how people's risk estimates respond to changes in the time of exposure to risk and found pervasive, systematic neglect for this temporal dimension. Community residents estimated that IRS tax audits were as likely within the next 2 years as 8 years; sexually active students viewed contracting an STD to be as likely within 1 year as 3 years. Even on-the-job construction workers exhibited temporal insensitivity for future job injuries. The author suggests that such insensitivity occurs because lay risk assessment is driven more by the impression a risk evokes than the quantitative aspects (e.g. time) it contains. Real-world consequences and possible debiasing mechanisms are discussed.
P2-43
Three Probabilistic Conceptualizations of Judgments and Preferences
Michel Regenwetter (McGill University)
We distinguish three fundamental probabilistic conceptualizations of preferences and/or judgments. These are 1) random relations, i.e., probability measures on m-ary relations; 2) random utilities, i.e., families of utility random variables; and 3: random functions, i.e., a probability measure over a function space. Although these three approaches usually involve different sample spaces, any of the three models can be translated into any of the others, under reasonable conditions.
P2-44
Affective Aspects of Decision Making
Alan Schwartz (Department of Psychology, UC Berkeley)
We experimentally investigate experiences associated with outcomes of choices between risky options. We present a theory of these experiences based on multiple counterfactual comparisons and examine the relationship between expected
feelings and decisions.
P2-45
Demographic Predictors Affecting Judgments of Child Abuse
David J. Weiss & Cynthia R. Deutsch (Department of Psychology, UCLA)
360 teachers each responded to a single vignette describing a hypothetical student in their classroom. The judgmental task called for estimation of the likelihood that the student was an abused child. The factors manipulated within the set of vignettes were the symptoms of abuse, the ethnicity of the child, the sex of the child, and the socioeconomic status of the family of the child. All of these have previously been implicated in the reporting of child abuse. Here, however, we asked only for recognition judgments. Results showed that subjects were more likely to respond positively only when the child displayed signs of physical injury. Because only one factor (symptoms of abuse) proved to be significant, no cognitive model of the judgmental process could be evaluated.
P2-46
Investigating the Determinants of Ambiguity Attitudes: Knowledge, Control, Accountability, and Decision Frame Kimberly
Most real-world decisions are made in the context of ambiguity, or uncertainty in probabilities. The common tendency towards ambiguity-aversion is well-documented. However, there is some evidence of ambiguity seeking behavior in certain circumstances. In this research, we begin to delineate under what circumstances ambiguity would be tolerated, or even preferred, and under what circumstances it will be avoided. The effects of the decision maker's knowledge, perceptions of control, and accountability on decisions involving ambiguous probabilities are explored experimentally. Further, how these factors are affected by whether the decision is framed as a gain or a loss is also examined.
P2-47
Systems-Thinking in an Intuitive Decision Making Environment: The Mexican Challenge
James L. Ritchie-Dunham (Graduate School of Business, University of Texas at Austin)
This paper proposes that Mexican organizations train their decision makers in high-tech, systems- thinking approaches instead of following the U.S. operations research legacy. The author proposes a "decision making technology - world vision" framework that demonstrates that it would be advantageous for Mexican organizations to adopt a high-tech, systems focus. Four case studies are summarized.
This form allows you to: (1) register for the November 2-4 annual meeting, (2) pay your 1997 dues, and (3) order two decision making journals for 1997 at a member discount rate. You may use the form for any one of these; but, doing all at once saves paperwork and should be more convenient for you.
NAME:
PHONE: ( )
FAX: ( )
E-MAIL:
ADDRESS:
CITY:
STATE/COUNTRY:
ZIP:
Address on label This is a
____ is correct ____ new address
MEMBER STUDENT* NON_MEMBER
Meeting registration fee $80.00 $40.00 $ 90.00
Late registration (After Oct. 22) $90.00 $45.00 $100.00
1997 Dues (including Newsletter) $20.00 $ 5.00
Past dues (please check label for
the last year for which you paid) ______ ______
Social Event: Second City (Sat., 11/2) $15.50 $15.50 $ 15.50
TOTAL $______ $______ $______
Note: Registration includes coffee breaks and reception and the Monday lunch. Dues are separate from registration fees. If you want to subscribe to either of the following journals for 1997, just check the appropriate space(s). Do NOT send your journal fees, you will be billed for the amount by the publishers. Please DO send your Society membership dues and meeting registration money.
I wish to subscribe to the following for 1997: [The journal will bill you later for the price of subscription at the special Society rate. The rates for 1996 are shown to give an indication of the approximate amounts.]
______ Organizational Behavior and Human Decision
Processes (12 issues, 1996 prices: $144 US &
Canada, $170 elsewhere)
______ Journal of Behavioral Decision Making (4 issues,
1996 price: $75)
Return this form with your check to:
Irwin Levin
Department of Psychology
University of Iowa
Iowa City, IA 52242
*Students must have the endorsement of a faculty member:
Faculty signature: Date:
Printed name: Institution: