Society for Computers in Psychology

Thirtieth Annual Meeting

November 16, 2000

Hyatt Regency New Orleans

President: Wally Beagley Secretary-Treasurer: Chris Wolfe
President-Elect: Sarah Ransdell Proceedings Editor: Jon Vaughan
Past President: John Krantz Federation Rep: Nancy Duncan
Webmaster: Bob Allan Program Chair: David Washburn
Steering Committee: George Wolford, Barry A. Tanner, Doug Eamon, Margaret Anderson, Gerardo Gonzalez, Richard Lehman, John Vokey, Curt Burgess, Chris Cozby


Greetings from the President...

Welcome to the 30th anniversary meeting of the Society for Computers in Psychology. Thanks to David Washburn for putting together a great program, and to all those who are sharing their work and ideas with us. A special welcome to those who are new to SCiP. If you are doing interesting work with computers, please consider making a presentation at next year's conference. (See the society's web site is at: www.lafayette.edu/~allanr/scip.html)...............................Wally Beagley, President

Greetings from the Program Chair,

It is my pleasure to welcome you to New Orleans and the 30th meeting of SCiP. At no time in history has the computer been so intricately a part of psychology. This statement is true not only for the ways in which we conduct ourselves as psychologists - the tools with which we collect our data, teach our classes, communicate with our colleagues, and even submit (with occasional frustrations like server failures!) our talks to professional meetings. Computer-directed behavior and computer-related cognition are also increasingly important as subject areas within our field. Consider that only one year ago the world was transfixed on the "Y2K bug" and its potential consequences - a fascination that was above all a psychological rather than a technological event.

The program for this year's meeting reflects the diverse ways in which computers are "in" contemporary psychology. I am excited to have Stevan Harnad at the meeting this year as our invited speaker, and look forward to the panel and audience discussion of the issues surrounding open archiving. I appreciate the contributions and patience of each of the program participants, as well as the efforts of the SCiP members who helped with reviews. I am particularly grateful for the assistance and direction of Wally Beagley and Chris Wolfe, who together with the Steering Committee keep the Society operational.

Finally, I challenge each of you to begin thinking already of outstanding contributions that you can make to the 31st annual program, and of colleagues who should also be participating in (and benefitting from) the Society. Your responsiveness in this area will ensure that next year's program chair enjoys the task as much as I did.....David Washburn, Program Chair

Session I 8:00 - 9:20 a.m. Regency F

COMPUTERS IN RESEARCH AND THEORY

CHAIR: Cyndi McDaniel

8:00 Hypermnesia and the Primitive Processes of Cognition

Gerald A. Epling, University of Texas at Dallas

epling@mindjava.com

The hypothesis that an item-specific encoding strategy can lead to observable increases in memory over time after learning (hypermnesia) as measured by recognition testing is explored. Hypermnesia was obtained following a specific incubation interval. This incubation period is considered to be a measure of the time complexity fo the human visual system involved in memory formation for the given stimuli.



8:20 Favored vs. surprising meaning activation in joke comprehension: A web-based investigation

Rachel G. Hull, Jyotsna Vaid, Francisco Martinez, & David P. Gerkens,

Texas A&M University

mrhull1@juno.com

What is the temporal course of activation for meanings elicited by the premise and punchline of joke texts? This study used a web-based design to introduce a semantic priming lexical decision paradigm with joke texts to determine the relative pattern of priming for target words. The target words were presented at different temporal points in the joke text and were related to either the initially favored interpretation or the surprising interpretation of each joke.



8:40 Psychological theorizing considered as reverse engineering

Thomas T. Hewett & Spiros Mancoridis

Drexel University

hewett@drexel.edu

This paper begins a conceptual exploration of the analogical correspondence between research and theorizing in psychology and the process of reverse engineering practiced by software engineers. First there is a generic characterization of reverse engineering and psychological research. Next the paper begins to explore the closeness of the analogy and describes some of the points at which the analogy does and does not hold.

9:00 Using Genetic Algorithms to Evolve Adaptive Strategies for Matrix Games

J. Neil Bearden

The University of North Carolina at Chapel Hill

Neil@unc.edu

Genetic algorithms were used to evolve adaptive (Q-learning) strategies for the iterated prisoner's dilemma and for other matrix games. The resulting strategies are more robust than deterministic strategies, which have been the focus of most of the research in this area (e.g., Axelrod, 1987), and are also more psychologically plausible. This methodology is computationally intensive and well illustrates a level of theorizing that is not possible without computers.







Session II 8:00 - 9:20 a.m. Regency G

AUTHORING AND EXPERIMENT GENERATION

CHAIR: Stanley H. Cohen, West Virginia University Cohen

8:00 XML Markup of Web Documents

David W. Stockburger

Southwest Missouri State University

dws148f@mail.smsu.edu

Extensible Markup Language (XML) provides considerable advantages for the web developer over Hypertext Markup Language (HTML). By marking a document for meaning and content rather than display, the web developer has a single source document that can be transformed into just about any current or future display form. The potential of XML markup will be demonstrated using an on-line statistics textbook.



8:20 Inquisit Programming to Demonstrate Cognitive Psychology Experiments

Scott Ottaway, Dale Dinnel, Jonathan Reed

Western Washington University and East Carolina University

ottaway@wwu.edu www.millisecond.com

Inquisit is an easy-to-use, highly-flexible scripting language for use in MS Windows environments. In addition to accurate timing and stimulus presentation, Inquisit allows researchers easily to incorporate graphics/images, sound presentation, and speech recognition into a research project. A brief overview of Inquisit will be provided, followed by a demonstration of some of its primary capabilities. Advantages relevant to on-going research programs, classroom demonstrations and data analysis will be discussed.

8:40 Real-time data collection in Linux: A case study

Steven A. Finney

Ohio State University

finney.17@osu.edu

Multi-user UNIX-like operating systems such as Linux are often considered unsuitable for real-time data collection because of the potential for indeterminate timing latencies resulting from preemptive scheduling. In this paper, I describe a data collection and auditory feedback manipulation system for tapping and music experiments that runs under Linux with reliable millisecond-level timing resolution. The implementation is discussed in detail, including programming methods, real-time support in standard Linux, and benchmarking procedures.



9:00 Using Package Files to extend and upgrade experiment generators such as E-Prime

Walter Schneider (University of Pittsburgh & Psychology Software Tools), Anthony Zuccolotto (Psychology Software Tools), & Brandon Cernicky (Psychology Software Tools)

walters@pstnet.com

Experiment generators often need to be extended to address specialized experimental needs such as biological recording, experimental tasks, and data base management. Package files provide subroutines with a E-Studio graphical interface, provide documentation to support use, and allow installation based automatic updating of previous programs. Examples of supporting brain imaging, specialized functions such as RSVP and up/down procedures, and data export for descriptive graphs are illustrated.



Session III 8:00 - 9:20 a.m. Regency H

TEACHING STATISTICS WITH COMPUTERS

CHAIR: David A. Washburn, Georgia State University

8:00 Evaluating Alternative Types Web-Based Activities in a Research Methods and Data Analysis Laboratory

Calvin P. Garbin, Marsha Bradley, Beryl Jean Mason, & John Orr (University of Nebraska-Lincoln)

cgarbin@unl.edu

The advantages of using web-based activities to augment lectures, as substitutes for lectures and to provide remedial instruction were examined. All three versions facilitated quiz and/or report writing performance, while providing for different uses of class time.

8:20 Difference to Inference: Teaching students to integrate deductive, inductive and statistical reasoning to make theoretical inferences through online interactivity.

Thomas E. Malloy and Gary C. Jensen

University of Utah

malloy@psych.utah.edu

www.utah.edu/stat/introstats www.psych.utah.edu/malloy/

Difference to Inference (www.utah.edu/stat/introstats) is a series of interrelated JAVA programs which use probability tools and two puzzle solving games to enable students actively to integrate statistical reasoning with deductive, inductive and inferential logic. Students must strategically plan a series of studies and then use the data from those studies to eliminate competing theories.



8:40 Teaching Anova with Spreadsheets

Thomy Nilsson

University of Prince Edward Island

nilsson@upei.ca

Most introductory statistics books still emphasize calculator methods, but few students will ever use calculators afterwards. Statistical programs such as SPSS provide little understanding of what happens. The otherwise tedious definition-based calculations using individual difference-from-the-mean squared are readily done on spreadsheets and show how the numbers work. Spreadsheet ANOVA's can be introduced with an even simpler method based on average deviations. Spreadsheets readily handle factorial and within- subjects designs too.



9:00 Utah Virtual Lab: JAVA Interactivity for teaching Science and Statistics Online

Thomas E. Malloy and Gary C. Jensen

University of Utah

malloy@psych.utah.edu

www.utah.edu/stat/introstats www.psych.utah.edu/malloy/

Utah Virtual Lab is a JAVA program run dynamically off a database. Instructors author a statistical virtual reality simulating theories and data in a research area by defining independent, predictor, and dependent variables and the relations among them. To discover the principles of this simulated reality, students go to a library, read theoretical overviews and scientific puzzles, and then go a lab, design a study, collect and analyze data, and write report.



Session IV 9:25 - 10:45 a.m. Regency F

COMPUTER UTILIZATION AND BROWSING BEHAVIOR

CHAIR: Wayne D. Gray , George Mason University

9:25 Qualitative Evaluation of Breast Cancer Web Sites: Which Is Better, Understandable Content or Maneuverability?

Patricia A. Marsh

Kansas State University

pmarsh@ksu.edu

Participants were instructed to find at least one example of a "useful" and "less than useful" breast cancer web site. Web addresses and written comments were compiled into response categories. Results from 130 participants highlight the continued importance of the maneuverability of a site in addition to it being informative.



9:45 Visualizing Individual Differences in Web Navigation: STRATDYN, a Tool for Analyzing Browsing and Search Patterns

Bettina Berendt & Elke Brenstein

Humboldt University

berendt@educat.hu-berlin.de

Navigational behavior during search and browsing sessions on the web is best analyzed with log file data. In this paper, we present a tool which aids in deriving meaningful quantitative and qualitative measures from server-generated log files, and provides easy-to-follow visualizations of navigational paths of individual users. With the help of STRATDYN we show, in two studies, that the effectiveness of search behavior is related to individual differences in the ability to concentrate.



10:05 How to Teach People About the Web: Image-Schematic Metaphor and the Construction of the Information Highway

Kimberly K. Barnard & Herbert L. Colston

University of Wisconsin - Parkside

barna000@uwp.edu

A popular metaphor used to describe the World-Wide-Web (web) has been that of an "Information Highway". Our study examined the use of other metaphors and analogy as expository tools in describing basic web components (i.e. mouse, modem, and routing server). It is suggested that the highly interactive and deep structural encoding nature of the image-schema makes the metaphor an effective tool for teaching web concepts.

10:25 Anatomy of a Birth: Analyzing Student Utilization of a Web-Assisted Course on "Psychology as a Profession"

Stanley H. Cohen and Lori M. Murray

West Virginia University

Scohen@wvu.edu

The present paper presents a preliminary assessment of student utilization of a course-assisted web site on psychology as a profession. The findings demonstrate the importance of understanding web utilization, especially as more courses in the psychology curriculum establish web sites. Assessment and evaluation of course materials provide an empirical basis for subsequent course and web site revision.





Session V 9:25 - 10:45 a.m. Regency G

COMPUTER RESOURCES FOR RESEARCH

CHAIR: William C. Schmidt, SUNY at Buffalo

9:25 Making Faces: Creating 3-Dimensional, Ecologically-Motivated Poseable Expressions

Jesse Spencer-Smith, Heather Wild, Ase Innes-Ker, Christy Duffy, Chad Edwards, Kristina Ervin, Jae Won Paik, Nicole Prause

Indiana University

jbspence@indiana.edu

Previous studies of the perception and processing of emotional expression in faces have been limited by the lack of realistic yet controlled stimuli. We created realistic computerized expressions, using 3D animation software, based on knowledge offacial musculature. The resultant stimuli are empirically validated by comparing intensity and accuracy ratings to those for real faces. With such stimuli, researchers can address opposing theories of emotion perception and processing.



9:45 cTRAIN: A Computer-aided Training System Developed in Supercard for Teaching Skills Using Behavioral Education Principles

W. Kent Anger, Diane S. Rohlman, Oregon Health Sciences University; David A. Eckerman, University of North Carolina; Ronald R. Reed, Ronald R. Reed Associates, John Kirkpatrick, Painters District Council #5

anger@ohsu.edu

A computer-aided training program was developed in SuperCard and piloted in painters using respiratory protection as an example of knowledge required in a blue collar occupation. Taking a modern programmed instruction approach, cTRAIN is structured as a series of "information sets" which consist of 3-5 (recommended) information "screens" followed by 1-3 (recommended) quiz screens structured as 4-response multiple choice questions. Data from painters support the technical choices of SuperCard and the MacOS with text-to-speech capability.



10:05 The RIT Wearable Eyetracker: Monitoring Eye Movements in Everyday Life

Jason Babcock and Jeff B. Pelz

Rochester Institute of Technology

pelz@cis.rit.edu

Visual perception must make sense of a field of view subtending over 10,000 deg^2, yet high-resolution vision is limited to a region subtending less than 1 deg^2. The mechanisms and strategies employed do not yield to introspection, and attempts to understand the process by piecemeal analysis of sub-systems have had only limited success. The RIT Wearable Eyetracker was designed to allow monitoring of subjects performing natural, 'everyday' activities outside of the laboratory.



10:25 Neuropsychological Applications of the Computerized Test System

David A. Washburn, Jonathan P. Gulledge, Georgia State University; William D. Hopkins, Berry College

The Computerized Test System has proven to be useful for studying cognition across primate species and for promoting the psychological well-being of these species. In this report, we report extensions of the CTS paradigm in directions that support further analysis of the brain-behavior relations that underlie task performance. Joystick use by rodents, the development of tasks relevant to neuropsychological assessment, and integration of the CTS with transcranial magnetic stimulation will be discussed.





Session VI 9:25 - 10:45 a.m. Regency H

STATISTICS & ANALYSIS

CHAIR: Thomy Nilsson, University of Prince Edward Island



9:25 Meta-Analysis Using Linear Mixed Models

Ching-Fan Sheu, DePaul University; Sawako Suzuki, University of California

sawako@uclink4.berkeley.edu

Psychologists often use special computer programs to perform meta-analysis. Until recently, this has been necessary because stanadard statistical packages do not provides procedures such analysis. The paper introduces the linear mixed models as a framework for meta-analysis in psychological research. This approach is illustrated with three examples using SAS PROC MIXED.



9:45 Using the General Processing Tree Model to Perform Simulations of Categorical Data

Jeffry S. Kellogg, Davidson College; Xiangen Hu, & William Marks, The University of Memphis

jeffkellogg@mail.com

This paper addresses the utility of the General Processing Tree (GPT) model computer program, GPT.EXE, which is specifically designed as a method for analyzing multinomial processing tree models based on the EM algorithm. After an explanation of how GPT.EXE can be used to conduct simulations, we present an example of using GPT.EXE to perform a simulation: a simulation of response bias in the process dissociation paradigm (Kellogg, Hu & Marks, under review).



10:05 Markov Chain Monte Carlo Estimation for General Processing Tree Models: A Program In S-PLUS

George Karabatsos

LSU Health Sciences Center

gkarab@lsumc.edu

The paper presentation will discuss a Markov-Chain Monte Carlo Estimation Algorithm that estimates parameters of the General Processing Tree (GPT) Models of cognitive psychology. The program for the algorithm is written in S-PLUS, and the user can estimate the parameters of any model in the GPT class. The estimation procedure is itself very simple, and therefore facilitates further developments in GPT model theory, particularly for GPT models that are rather complex.

10:25 A Nonlinear Regression Approach to Estimating Signal Detection Models for Rating Data

Ching-Fan Sheu, DePaul University; Andrew Heathcote, The University of Newcastle, Australia

csheu@depaul.edu

This paper considers general nonlinear regression methodology for receiver operating characteristic (ROC) curve estimation. This approach has three advantages: First, it links statistical modeling for ordinal data to the practice of ROC data analysis in experimental psychology. Second, it requires only standard statistical software such as SAS for computations. Third, it greatly facilitates the ability to explore effects of covariates on model parameters. The approach is illustrated using two real data examples from rating-ROC experiments.











Session VII 10:50 a.m. - NOON Regency F

Presidential Address

Chair: John Krantz, Hanover College

10:50 Where Has All the Software Gone? Why We Need More Psychologist/Programmers

Walter Beagley

Alma College

The talk will include a demonstration of EL Knife, a data utility for reconfiguring spreadsheets, available for free download from http://www.alma.edu/el


To be followed immediately by the Castellan Student Award presentation

and the annual Business Meeting (open meeting, all invited).

Minutes of Business Meeting, Thursday, Nov. 18, 1999. 12:45pm. Century Plaza Hotel, Los Angeles, CA.
1. President Wally Beagley convened the meeting. Thanks to program co-chairs John Vokey and Scott Allen. David Washburn from Georgia State University will be the 2000 program chair.
2. Program Chair Report (Scott Allen): The Castellan Award was presented to Patrick Conley (UC Riverside). 102 people registered for the meeting. There are 51 papers and 2 symposia.
3. Secretary-Treasurer Report (Chris Cozby): Cozby thanked the vendors who placed ads in the program and have display information available. The society has a cash balance of $15,889.58. Sarah Ransdell is the new President-Elect and Chris Wolfe is the new Secretary-Treasurer. Steering committee members elected were Doug Eamon, Cozby, John Vokey, and Curt Burgess. The 2000 meeting will be held in New Orleans.
4. Proceedings Report (Jon Vaughan): All submissions of SCiP papers for publication in BRMIC are undergoing review.
5. Federation Report (David Johnson): Johnson described the proposal to fund electronic archiving of research data in the social and behavioral sciences.
Meeting adjourned at 1pm.




Session VIII 1:00 - 2:00 p.m. Regency F

PANEL DISCUSSION: USE OF ON-LINE ACTIVITIES FOR ACADEMIC ADVANCEMENT

CHAIR: Margaret D. Anderson, SUNY at Cortland

Participants: Doug Eamon, University of Wisconsin Whitewater

John Krantz, Hanover College

Sarah Ransdell, Florida Atlantic University

Ulf Reips, Universitat Zurich

Margaret D. Anderson, SUNY at Cortland

ANDERSMD@SNYCORVA.CORTLAND.EDU

This panel discussion will focus on policies various institutions are implementing for the evaluation of on-line activities for academic advancement. Specific issues such as: submission of materials to on-line journals; editorship of on-line sites; creation/maintenance of academic web sites; on-line research; participation in on-line instruction or conferences; and evaluation of on-line courses, will be discussed. Panelists represent a wide range of US and international Universities and Colleges and will provide a diversity of models currently in use or under construction. In addition suggestions will be put forward concerning possible avenues individuals may follow if they wish to see their on-line activities contribute to furthering their academic careers.







Session IX 1:00 - 2:00 p.m. Regency G

PRESENTATION AND TIMING

CHAIR: Scott Ottaway, Western Washington University

1:00 Presentation Accuracy of Web Animation Methods

William C. Schmidt

SUNY at Buffalo

wcswcs@acsu.buffalo.edu http://or.psychology.dal.ca/~wcs

Vision experiments on the web often require precise timing in the delivery of stimuli. Several animation methods were assessed, including those requiring programming and browser extensions (Java, Javascript/JScript, DHTML and Authorware) as well as those that do not (GIF, PNG). The difference between timing stipulated in the stimulus generation versus that obtained across several browser versions, operating system platforms and speed of computer system is reported and ways to decrease this discrepancy are presented.

1:20 Reaction Time Measurement by JAVA-Applets Implementing Internet-Based Experiments

Jan Eichstaedt

Universität der Bundeswehr Hamburg

jan.eichstaedt@UniBw-Hamburg.DE

Without precaution JAVA-applets are too imprecise when measuring reaction time. The present study introduces a new technique that identifies inaccurate reaction time measurements by testing whether the client-side JAVA-implementation makes false measurements of a time interval of known length. The results show the effectiveness of the new technique and indicate that it enables applets to measure reaction time when conducting experiments in the Internet.



1:40 Millisecond Timing on PCs and Macs

Joseph MacInnes and Tracy Taylor

Dalhousie University

macinnwj@cs.dal.ca

Modern computer operating systems have many benefits and pitfalls for experimenters looking for real time platforms. 32 bit graphics and video acceleration have drastically improved the quality of experiments that can be displayed on new computers; however, improved multi-tasking has made it difficulty to achieve a real-time environment. The capability of many PC and Mac Operating systems to achieve true real-time will be demonstrated, as well as the problems in assuming that they will accomplish real time automatically.





Session X 1:00 - 2:00 p.m. Regency H

THE WEB AS A TOOL FOR THE ACADEMY

CHAIR: Gerald A. Epling, University of Texas at Dallas

1:00 Time As A Variable In Learning

Roman Taraban

Texas Tech University

r.taraban@ttu.edu

Learning takes time. Support for this commonsense notion was found in three replications of an upper-level undergraduate course in which a significant portion of the material and activities were delivered over the WWW. Computer records were correlated with in-class test performance and showed significant positive correlations in a majority of the analyses. The data revealed astonishingly low levels of engagement with course materials and an ill-adapted pattern of study time distribution.

1:20 The Psychology Graduate Applicant's Portal

Curt Burgess, Patrick Conley, & Catherine Decker

University of California, Riverside

Email: curt@citrus.ucr.edu www.psychgrad.org

The goal of this project is to create a clearing house of information on applying to graduate school in psychology, and to make this knowledge easily accessible on a single website. This website, called the Psychology Graduate Applicant's Portal, contains targeted links to other webpages with valuable information, book recommendations on the application process, polls to improve the site, site-unique advice on graduate programs, a message board to exchange information and experiences, and other information relevant to the applications process.

1:40 A Proposal for the Development of a Stimulus Bank

Brian MacWhinney

Carnegie Mellon University

macw@cmu.edu

A series of new technical advances make it possible to construct an international web-accessible database for materials and procedures in Experimental Psychology. This can be done in the context of the new NSF TalkBank initiative (http://talkbank.org) that is funding the construction of a system for streaming video linked to transcription. This facility would be particularly well suited to the archiving of experimental methods in studies of perception, motion, and problem-solving. Concrete examples will be presented.





Session XI 2:05 - 3:45 p.m. Regency G

SOFTWARE SOLUTIONS

CHAIR: Bettina Berendt, Humboldt University

2:05 VIDANN, a Video Annotation System

A. De Clercq, A. Buysse, H. Roeyers, K. Ponnet, L. Verhofstadt, W. Ickes

Ghent University (Belgium), University of Texas at Arlington

armand.declercq@rug.ac.be

VIDANN is a programme that aims at computerizing the empathic accuracy design (Ickes et al., 1990). It allows participants to start and stop a video and to write their annotations on paper attached to a writing tablet. The system is designed as a Microsoft ActiveX module. It can be adapted by the experimenter through a VBScript. All data, including the handwriting is stored in an XML database. A Wizard is designed for generating VBScripts for standard configurations.

2:25 Argus -- A Flexible, Multiple-Cue Probability Task for Modeling and Studying Individual and Team Cognitive Workload, Strategies, and Microstrategies

Wayne D. Gray and Michael J. Schoelles

George Mason University

gray@gmu.edu

Argus simulates a radar-like target classification task. It was developed to support research in measuring and modeling cognitive workload. Argus has been used in both single-subject and team mode. Features of Argus include flexible experimenter control over cognitive workload, as well as extensive data collection and data playback facilities. In addition, models written in ACT-R/PM (Byrne & Anderson, 1998; Byrne 1999) interact with Argus using the same interface as do human subjects.



2:45 ACT-PRO Action Protocol Analyzer: A Tool for Analyzing Discrete Action Protocols

Wai-Tat Fu

George Mason University

wfu@gmu.edu

Action protocols analysis often entails matching data to patterns specified by the researchers. An ACTion PROtocol analyzer (ACT-PRO) was developed that (1) groups action sequences into labeled strings, and (2) matches these labeled strings to a hierarchical pattern. ACT-PRO can be used to determine how well the data fit the patterns, and focus the experimenter's attention on data that do not fit the pre-specified patterns.



3:05 RatCog: A GUI maze Simulation Tool with Plugin "Rat Brains"

Christopher G. Prince, University of Minnesota Duluth; John Talton, Louisiana Tech University; István S. N. Berkeley, & Cengiz Gunay, University of Lousiana Lafayette

chris@cprince.com

We have implemented RatCog, a Graphical User Interface (GUI) radial-maze simulation tool providing various computational models of rats. Rat models are loaded as runtime plugin files, and an Application Programming Interface (API) enables additional plugins to be created. One implemented plugin is a back-propagation trained connectionist model. GUI features include maze graphics and performance statistics. The GUI makes it easier to use these computional models, while the plugins make the models widely available.

3:25 The Decision Simulator P1198: a Software Program to Trace Decision Processes in Loan Decisions to Small Business Enterprises

Patric Andersson

Center for Economic Psychology, Stockholm School of Economics

ppa@hhs.se

This paper describes a process-tracing program designed to capture information acquisition behavior in lending. The program has two interacting systems: a database and a user interface. The database includes three realistic loan applications. The user interface consists of different but interacting screens that enable the subject to operate the software. Log-files register the information acquisition behavior by the subject. Some empirical findings are presented to show the software's potentials in providing insights into decision processes in lending.





Session XII 2:05 - 3:45 p.m. Regency H

RESEARCH ON THE WEB

CHAIR: Dalila Bovet, Georgia State University

2:05 Effects of Dropout Rate and Efforts to Reduce it in one Web-based

Kevin M. O'Neil & Steven D. Penrod

University of Nebraska-Lincoln

koneil1@bigred.unl.edu

Web-based researchers are in a position to measure and control dropout rates. Although there were no main effects in the present study of whether participants were asked to enter an email address, were offered payment through a lottery, or of day of the week that participants completed the experiment, there were several two- and three-way interactions between these variables, the manipulated variables, and attitudes.



2:25 Interactive Visual Speech Experiments Presented via the Internet

Michael S. Gordon & Lawrence D. Rosenblum

University of California, Riverside

Rosenblu@citrus.ucr.edu

Internet resources now enable laboratories to make experiments fully available online. To take advantage of these new technologies we have several interactive experiments that can be accessed via the internet. These experiments explore visual speech and face perception using software that can be scripted by intermediate level computer users. This website allows internet users to participate in short experiments and generate real data that they can access for their own purposes.

2:45 Investigating the Validity of a New Method for Jury Research: Imposing the Death Penalty over the World-wide Web

Steven D. Penrod and Kevin M. O'Neil

University of Nebraska-Lincoln

koneil1@bigred.unl.edu

Web-based research applies a relatively new method for collecting data and may involve different samples than prior research. Within the research concerning jury behavior, infrequently is there any reported mean differences nor any interactions between the type of sample or the method and manipulated variables. Three studies investigated differences between using the Web as compared to pencil-and-paper or between undergraduate and Web-based adult samples. Results revealed no method differences but an occasional sample difference.

3:05 Evaluating Differences in Self-Disclosure Based on Methods of Response: Internet Response vs. Paper-and-Pencil Response

Kristine Schuster, Ian Birky, and Bruce Sharkin

Lehigh University

krs3@lehigh.edu

Only recently has the study of the validity of Internet (Web) based research been considered. Many researchers have found that the results of web-based research are comparable to the results of studies conducted in more traditional ways, with one exception being that self-disclosure is often greater in research conducted via the Internet. We present data that contradicts previous findings and suggest potential causes of disparity between past and present findings.



3:25 The Web Experimental Psychology Lab: Five Years of Data Collection on the Internet

Ulf-Dietrich Reips

Universität Zurich

ureips@genpsy.unizh.ch

www.psych.unizh.ch/genpsy/

www.psych.unizh.ch/genpsy/Ulf/Lab/WebExpPsyLab.html

In fall 1995 the worldwide accessible Web Experimental Psychology Lab (http://www.genpsylab.unizh.ch) opened its doors to Web surfers and Web experimenters. It offers a highly frequented place to conduct true experiments over the Internet. Data from five years of laboratory running time are presented along with recommendations for setting up and maintaining a virtual laboratory, including sections on data security, data quality, administration, design, software and hardware, marketing, other Web laboratories, and ethical issues.



Session XIII 2:05 - 3:45 p.m. Regency F

TEACHING ON THE WEB

CHAIR: Roman Taraban, Texas Tech University

2:05 Mastery Quizzes on the Internet: Results from a Web-based Introductory Psychology Course

William S. Maki & Ruth H. Maki

Texas Tech University

Bill.Maki@ttu.edu

QUEB is a system for delivery of multiple choice quizzes over the web implemented as a set of Perl scripts. Items are contained in a text file, are randomly selected and ordered, and rich feedback is provided to the student. The system has been in use for four years, is highly rated by students, and is correlated with performance on examinations



2:25 System for the Teaching of Experimental Psychology (STEP)

Brian MacWhinney, Carnegie Mellon University; Ping Li, University of Richmond

macw@cmu.edu

In order to learn what it means to conduct empirical research in Psychology, students need to design and analyze their own experiments. Using the new E-Prime experiment development system, we have now constructed a set of 50 classic demonstration experiments that are publicly available for running and downloading at http://step.psy.cmu.edu. We will demonstrate 12 of these experiments, including 6 that were built by undergraduates in Experimental Research Methods.

2:45 A Website for an Interdisciplinary Course in Cognitive Science

Paula Goolkasian and Lori Van Wallendael

University of North Carolina, Charlotte

pagoolka@email.uncc.edu

This paper describes the development and evaluation of a Website (http://web2-pc.uncc.edu/cogsci) to support an interdisciplinary course in cognitive science. The modules include introductory reading material, interactive exercises/virtual laboratory, and pointers to existing material on the web. Our approach to using the web in support of instruction differs from distance learning initiatives because it is centered on an instructor and classroom experiences. The web-based modules are used to supplement classroom lectures and to provide an interdisciplinary perspective.

3:05 Using On-Line and CD-ROM Database Archives as an Adjunct to Teaching Survey Research Methods

Doug Eamon

University of Wisconsin-Whitewater

eamond@uwwvax.uww.edu http://facstaff.uww.edu/eamond

This paper examines the use of on-line and CD-ROM databases as an alternative to student designed survey instruments. The availability of large amounts of raw data from which the student may select variables to be analyzed provides unique opportunities for teaching research methods. The use of these databases in the author's research methods classes is reported. The recent initiative forming a consortium of interested groups is expected to provide expanded opportunities and added impetus for teaching database techniques.



3:25 Using Student-Authored Web Pages to Promote Active Learning, Student Performance and Increased Use of the Internet

Calvin P. Garbin, University of Nebraska - Lincoln; & Katherine E. Brooks, University of New Orleans

cgarbin@unl.edu

The explosion of web-based applications for teaching Psychology has included little attention to the educational benefits of student-authored materials. Students rated authoring, peer-reviewing and reading other students' webpages as more helpful than reading the textbook and exam scores showed the same pattern. Students who reviewed webpages before lectures rated those lectures as "better" and "more helpful". Students reported becoming more comfortable with the Internet and increasing its unassigned educational use.











Session XIV 3:50 - 5:30 p.m. Regency F

INVITED ADDRESS AND PANEL DISCUSSION

CHAIR: Wally Beagley, Alma College



3:50 For Whom the Gate Tolls? How to Free the Refereed Literature Through Auto-Archiving, Now.

Stevan Harnad

University of Southampton

CLICK HERE FOR THE ABSTRACT

4:40 Panel Discussion

John E. Belquist, Psychonomic Society Publications

Arthur B. Markman, University of Texas

Hal Warren, American Psychological Association

Audience discussion

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

From 5:30 to 7:00 pm, there will be demonstrations and symposia in Regency G and H

Regency G: E-Prime - A cross platform Experiment Generator Studio for Computerized Behavioral Research and Teaching (Organized by Walter Schneider)

Regency H: PsychExperiments (Organized by Ken McGraw)







30th Annual Meeting of the

Society for Computers

in Psychology























Hyatt Regency Hotel

New Orleans, Louisiana

November 16, 2000