心理学计算机化实验的形式与生成系统

文章先概述单机实验、多种设备联机实验和在线实验三种心理学计算机化实验方法的特点,然后对E-Prime等四种基于Windows的心理学计算机化实验生成系统进行介绍,最后对心理学计算机化实验在形式、技术与应用上的未来进行展望。     

A comparison of instruction formats for administering a computerized behavioral test

Shaping and fading techniques were employed to replace written instructions used in computerized behavioral tests. Apple Macintosh computers and HyperCard with Pascal extensions were chosen as the development environment to implement the new techniques. Initial Findings demonstrate the effectiveness and relative efficiency of shaping techniques to teach correct performance on the Symbol-Digit Test.     

Experiments in Group Decision Making Communication Process and Outcome in Face-to-Face Versus Computerized Conferences

A 2 × 2 factorial design was used to explore the process and outcome of small group problem-solving discussions for two modes of communication (face-to-face and computerized conferencing) and two types of tasks (a qualitative human relations task and a scientific ranking test with a criterion solution). Interaction process was coded using Bales Interaction Process Analysis. There were two to three times as many communication units in the face to-face groups consisting of five members each as in the computerized conferencing mode of communication during the same elapsed time. Group decisions were equally good in the two modes, but the groups were less likely to reach agreement in the computerized conferencing mode. There were proportionately more of the types of task-oriented communication associated with decision quality in the computerized conferences.     

DEVELOPMENT AND VALIDATION OF A COMPUTERIZED INTERPRETATION SYSTEM FOR PERSONNEL TESTS

A computerized system was developed for generating narrative interpretations of scores from a battery of personnel screening tests. The report structure and interpretive statement library were designed to capture the test expertise and interpretive strategies of a panel of testing experts. This was accomplished by enumerating the questions that the experts believed the battery could answer, developing answers to these questions, and devising rules for selecting the appropriate answers based on test-battery scores. The accuracy, thoroughness, readability, and coherence of the computer-generated reports were evaluated in comparison to reports generated by human experts for the same examinees. Results of the evaluation showed the computerized reports to be more accurate and thorough, as readable, and somewhat less coherent than interpretations generated by the typical human expert. The computerized system development and validation strategies described are useful for other applications in which numbers are interpreted in a narrative report format.     

Item type,occlusion, and gender differences in mental rotation

Two experiments were conducted to examine the role of item type in mental rotation. In each experiment, participants completed two computerized mental rotation tasks, one with blocks as stimuli and one with human figures as stimuli. The tasks were formatted either as a multiple-choice psychometric test (Experiment 1) or as a same–different type task (Experiment 2). Aside from the expected replication of a decreased effect of occlusion on women's accuracy when processing human figures compared to block figures, it was hypothesized that response times would increase when processing the complex but familiar human figures, compared to the simple but unfamiliar block figures. In Experiment 1, the results relevant to occlusion were replicated. However, the presence of a speed–accuracy trade-off suggested that participants processed human figures faster but less accurately than block figures. In Experiment 2, both men and women performed faster and more accurately when processing occluded human figures than when processing nonoccluded human figures. The effect of item type, its potential link to embodied cognition, and the role of strategy selection on gender differences in mental rotation are discussed.     

IGT-Open: An open-source,computerized version of the Iowa Gambling Task

The Iowa Gambling Task (IGT) is commonly used to understand the processes involved in decision-making. Though the task was originally run without a computer, using a computerized version of the task has become typical. These computerized versions of the IGT are useful, because they can make the task more standardized across studies and allow for the task to be used in environments where a physical version of the task may be difficult or impossible to use (e.g., while collecting brain imaging data). Though these computerized versions of the IGT have been useful for experimentation, having multiple software implementations of the task could present reliability issues. We present an open-source software version of the Iowa Gambling Task (called IGT-Open) that allows for millisecond visual presentation accuracy and is freely available to be used and modified. This software has been used to collect data from human subjects and also has been used to run model-based simulations with computational process models developed to run in the ACT-R architecture.     

Taskomat: Evaluation of a Computerized Test Battery

A case study on a computerized test battery is presented based on more than a decade of use in military and civil settings. The study sheds light on the general issues associated with computerized testing. The accent is on the contrast between computerized and traditional paper-and-pencil testing. The computerized battery, called Taskomat, is founded on concepts arising from human performance theory. The tests of the battery are described, as well as issues related to the hardware (cost of purchasing, hardware reliability and industrial styling). Criteria used to evaluate the content of the tests were: reliability, robustness of effects, uniqueness or newness of abilities relative to paper-and-pencil testing, and validity of these abilities for predicting occupational success. A conclusion is that a theoretical foundation helps to increase the likelihood of assessing new abilities.     

A procedure for the computerized representation and presentation of rotated and reflected stimuli

Psychological experiments in mental rotation and pattern recognition often require the presentation of stimuli in a variety of different orientations. Traditionally these stimuli have been presented with a tachistoscope; however, there are several advantages to using a system that presents such stimuli on a microcomputer. Unfortunately, in a typical experiment of this type, stimuli are presented after several different rotations, reflections, and translations, making the computerized representation of all the different permutations time- and memory-consuming. This paper describes a method of representation, utilizing a modified polar coordinate system, which allows for the efficient storage and transformation of such stimuli.     

Computerized Testing in the United States Air Force

Computerized testing in the United States Air Force finds its historical beginnings in apparatus tests developed during World War II. The advent of fast, small, and inexpensive computers led to the establishment of a research program to use computers in pilot selection. Later, computerized tests were demonstrated for more familiar aptitude measures such as verbal and quantitative ability. Several large-scale efforts to implement computerized testing have proved informative. Many Air Force tests are routinely administered by computer. Continued advances in computer technology promise to make computerized testing more common and more commonplace. Computer use in measuring physiological correlates of human ability is discussed.     

PsychMate: Providing psychology majors the tools to do real experiments and learn empirical methods

PsychMate is a set of software tools for undergraduate psychology students to run, develop, and analyze computerized experiments. It includes 30 psychological experiments in the areas of perception, cognition, social psychology, human factors, and cognitive neuroscience. Students run experiments themselves and see basic results immediately. The automatic spreadsheet analysis forms allow them to aggregate data and create analyses, presentations, and Web pages with a single click. Students can use the Psychology Experiment Authoring Kit experiment editor to create their own experiments in minutes and run experiments with other students using Web-based experiment-management tools. The Brain-Tutor and BrainViewer applications teach brain anatomy and permit students to analyze fMRI brain imaging data from subjects who have performed the same memory experiments in which they participated. PsychMate has been used in 83 classes in which 1,533 students submitted 5,464 completed experiments with few (less than 1%) requests for help and a very positive rating of the research experience.     

The Effect of a Computerized Decision Aid on the Development of Knowledge

This study provides evidence on how the use of a computerized decision aid affects knowledge acquisition. While the intent of such computer-based decision aids is to improve and support unaided human judgments, it is unclear whether the use of the aids actually facilitate knowledge transfer. Seventy-six subjects completed a hypothetical decision task where one of two types of decision aids was used: an expert system or traditional textual materials. Results suggest that those using a computer-based decision aid acquired and encoded in memory less declarative knowledge than those using textual reference materials. Implications of the findings are discussed.     

心理实验系统E-Prime介绍及其应用

心理实验正在进一步计算机化,文章介绍了这种发展趋势,概述了心理实验系统的发展,并重点介绍了标准化实验系统E-Prime。文中概述了E-Prime的功能及其与DMDX的比较,介绍了E-Prime作为实验教学系统和实验生成系统在实验教学和实验研究的应用,讨论了E-Prime应用中的一些问胚。     

A Comparison of Two Process Tracing Methods for Choice Tasks

Process tracing methods, particularly those based on information acquisition, are becoming commonplace. Because of this, it is important to examine both the reactivity and the validity of these techniques. This research compares information acquisition behavior for choice tasks using Mouselab, a computerized process tracing tool, and Eyegaze, an eye tracking system. In an experiment using apartment selection tasks and gambles, we found significant differences contingent upon the process tracing method for 10 process tracing measures including subsequent choices. Computerized process tracing tools increase the amount of time needed to acquire information compared with eye tracking equipment. As a result, subjects using Mouselab tend to have more systematic information acquisition behavior than that observed with eye tracking equipment. Additional research is needed to explore the magnitude and consequences of these differences.     

Definitions as theories of word meaning

Dictionary definitions are often used in education and in computerized databases for processing natural language, but their accuracy as models of word meaning has not been carefully evaluated. An experiment involving fourth graders and adults measured the efficacy of dictionary definitions in marking distinctions in sense and found them to be very poor for children, but fairly helpful for adults. A second experiment compared dictionary definitions to definitions written by college students who had sorted sets of sentence citations, and found both moderately effective in teaching word use. Good definitions were produced less often for high-frequency words than for low-frequency words, given equivalent numbers of citations, suggesting that sentences using high-frequency words are less informative overall. Also, the more citations available to the definition writer, the better the definition was in teaching word use, suggesting that information about word meaning is cumulative and describable.This research was partially supported by grants from the Spencer and Sloan Foundations to George A. Miller.     

Real-time generation of representations for cognitive models

Computerized models of human cognition and behavior have been used to help researchers refine their theories of cognitive processes. Two difficulties inherent in using these cognitive models are that building representations of to-be-modeled groups or individuals is (1) time-consuming and (2) prone to error when performed manually. The use of computerized tools to build representations for cognitive models could alleviate these problems. Such a set of tools has been developed and applied to the task of modeling individual aviation pilots performing flight maneuvers in real time. A study conducted with 27 pilots confirms the accuracy of the computer-generated representations.     

Assessing short summaries with human judgments procedure and latent semantic analysis in narrative and expository texts

In the present study, we tested a computer-based procedure for assessing very concise summaries (50 words long) of two types of text (narrative and expository) using latent semantic analysis (LSA) in comparison with the judgments of four human experts. LSA was used to estimate semantic similarity using six different methods: four holistic (summary-text, summary-summaries, summary-expert summaries, and pregraded-ungraded summary) and two componential (summary-sentence text and summary-main sentence text). A total of 390 Spanish middle and high school students (14–16 years old) and six experts read a narrative or expository text and later summarized it. The results support the viability of developing a computerized assessment tool using human judgments and LSA, although the correlation between human judgments and LSA was higher in the narrative text than in the expository, and LSA correlated more with human content ratings than with human coherence ratings. Finally, the holistic methods were found to be more reliable than the componential methods analyzed in this study.     

Can AI language models replace human participants?

Recent work suggests that language models such as GPT can make human-like judgments across a number of domains. We explore whether and when language models might replace human participants in psychological science. We review nascent research, provide a theoretical model, and outline caveats of using AI as a participant.