New tool makes online research easier

A web site enables researchers to conduct, participate in and even create their own online psychology experiments for free.

By Bridget Murray
Monitor staff

First, Ken McGraw?s students had to wait their turn to use the psychology laboratory?s dinosaur of a computer. For some, that meant waiting until Friday afternoons after 4. When it was finally their turn, they?d operate it in teams of two because a lack of money for undergraduate research equipment?typical of most universities?meant they were all relegated to one hand-me-down desktop PC.

As they input information, the ancient machine would languidly chew on the data, spluttering and seizing up at unpredictable intervals. Students in McGraw?s undergraduate lab classes endured this rigamarole for years. But they need suffer it no longer.

With the new Internet site, PsychExps (www.olemiss.edu/psychexps/), developed by McGraw, electrical engineer Mark Tew, PhD, and graduate student John Williams, students can speedily perform experimental tasks right on the World Wide Web. And they can do it night or day, in the comfort of their home or dorm room. If their class is running an experiment on word recognition, for instance, the web site records each student?s performance data and compiles the whole class?s results. Friends and other student volunteers can participate in the experiments too.

And University of Mississippi students aren?t the only ones benefiting from the innovation.

Using the web site, students and researchers throughout the country and across the globe can conduct, participate in and even create their own online psychology experiments, for free.

According to McGraw, schools that don?t have the funds to offer lab courses now can.

'The beauty of this technology is that we can deliver the same experience on the web that people get using programs written and compiled to run locally,' says McGraw. 'And instead of collecting data sequentially, data can be collected simultaneously from as many people as choose to log on.

'This is a real benefit to students who are trying to cram an independent research project, like a senior thesis, into a single 15-week semester. Rather than spend the bulk of their time on data collection, they can spend it where it matters?on analysis and writing their report.'

The PsychExps web site was born of McGraw?s dissatisfaction with software used for collecting data in undergraduate research classes. 'Basically, commercial software allows you to conduct somebody else?s experiment, not your own,' McGraw says. Then he discovered via Tew that it?s relatively easy to use multimedia authoring tools, in particular Macromedia?s Authorware?, to develop and deliver programs on the web. He and Tew decided to use the user-friendly tool to craft and post a series of web-based psychology experiments for everybody to use. McGraw and Tew were not the first to discover the benefits of Authorware as a programming tool for psychology experiments.

A web site created by psychologists David Eckerman, PhD, of the University of North Carolina?Chapel Hill and Dan Ariely, PhD, now at the Massachusetts Institute of Technology, offers Authorware programs for psychology experiments, but the programs cannot be run at the site. Their site is called ?The Class?: Experimental Psychology (web.mit.edu/ariely/www/psychlab/).

McGraw and Tew also weren?t the first to create web-based interactive experiences that add a realistic spin to textbook explanations of phenomena such as visual illusions or brain hemisphere differences. A number of sites offer such demonstrations. For example, Gary Bradshaw, PhD, at the University of Illinois, pioneered Internet Psychology Lab (kahuna.psych.uiuc.edu/ipl), which presents online demonstrations in the areas of cognition, memory and learning, and visual and auditory perception. Bradshaw explains, however, that his site is meant for exploratory learning, not data collection.

Because their idea was novel, McGraw and Tew secured a three-year grant for more than $200,000 from the federal government?s Fund for the Improvement of Postsecondary Education (FIPSE) and close to $70,000 from local sources. With their funding in hand and numerous late-night planning meetings behind them, McGraw, Williams and Tew unveiled PsychExps.

What it can do

To tap into the experiments offered on PsychExps, McGraw advises newcomers to the site to first click on the ?What You Need? button at the left of your screen. This tells you if you have the required software for running the experiments: Macromedia?s Shockwave plug-in for Authorware. People who don?t have Shockwave can download it via a provided link. Once it?s loaded, users can tap their choice of several basic psychology experiments, including ones on facial recognition, mental rotation, temporal judgment and memory of musical pitches. Click on the pitches experiment, for example, and you discern the similarity of various musical tones. McGraw began using the experiments with his undergraduates this fall. But University of Indiana psychology instructor David Huber was the first to try the site with 24 students last summer. Huber required 24 students to try several tasks with their friends, among them the Stroop Divided Field Experiment. (The Stroop task measures how long it takes people to name the color in which a word is printed, such as ?red,? when the word identifies a different color, such as blue.) The students formed hypotheses, tabulated their data and watched the results unfold, says Huber. ?They felt ownership of the experiment because they did all the data gathering and analyses themselves, and they didn?t know what the results would be,? says Huber. Some were disappointed when data from their own subjects didn?t support the class?s hypotheses, but generally the class averages supported the hypotheses, he says.

Some students also experienced technological glitches using the web site. ?But even that was probably a good thing because they gained experience fixing computer problems,? Huber says.

What it means for research

As much as McGraw and Tew have geared the site for undergraduate lab classes, they claim its research potential reaches much further than the classroom. One of McGraw?s undergraduate honor students is using the PsychExps site?s Stroop task to study brain hemisphere differences among people with a form of mild mental retardation known as Prader-Willi Syndrome.

The student, Shari-Ann Gregoire, visited the site of her research, the Northern Mississippi Regional Center in Oxford, several times to supervise eight Prader-Willi patients as they performed the task. Gregoire then downloaded the data, analyzed it and compared it with data from normal controls. She plans to publish the results in an undergraduate research journal.

And, says McGraw, the PsychExps site?s universal reach allows researchers like Gregoire to collect data from people living far away. Gregoire, for example, could extend her sample with Prader-Willi patients in other states. She says she may pursue that larger project in graduate school.

Indeed, Tew believes the PsychExps site?s greatest lure for scientists is its ability to handle large research data sets from widespread sources. ?Theirs is a site for hard-core experimental work,? agrees Bradshaw.

The only major concern Tew has about the site is how reliably it measures reaction time across different computing environments, an issue he and McGraw are researching through quantitative comparisons. Both welcome criticism and suggestions for improving the site, preferring to think of it as a ?work in progress.? Ultimately, when the FIPSE grant expires, they hope to find an organization that will continue managing and updating the site for a wide network of users.

?We don?t want this to be a book on a shelf but a useful living thing,? says McGraw. ?We hope people will modify our programs and add their own. We want it to be everybody?s web site.? Toward that end, they plan to work one-on-one and in workshops with those wishing to help develop PsychExps. ?Please tell people to contact us [through the site],? McGraw says.