Molar View of Project Implicit Infrastructure

Is current?: 
Yes

As a research enterprise, the Project Implicit infrastructure is comprised of many components.  These components can be conceptualized as two interactive halves - one being the web presence for participants to complete studies, and the other being the core infrastructure that is managed by administrators, developed by programmers, and used by researchers to build and host studies and to collect and store study data. 

Project Implicit from the participant perspective

Participants engage with the Project Implicit infrastructure in a few different ways.  The most prominent access point is https://implicit.harvard.edu/.  At this public entry point, participants can choose to visit the Demonstration site, International sites, Research site, or (starting in 2011) PIMH - Project Implicit Mental Health.  On some occasions, a Featured Task is accessible directly from this main page. 

The Demonstration site opened in the Fall of 1998 and provides 10-minute, demonstration studies of 14 topics relating to social group attitudes and stereotypes.  Participants select which topic they would like to complete, and can do as many as they like.  The International sites are replications of the Demonstration site that are revised for a particular national context.  For most, this means that the questions and some of the demonstration tests are revised to be relevant in that culture.  And, for many, the site materials are translated into the dominant language of that nation.  There are more than 30 international sites in 22 different languages.  Most international sites have about 6 of the demonstration studies.

The Research site opened in 2003.  There, participants must register an account and provide an email address as the login identifier.  During registration, participants complete a short demographics questionnaire.  Each time a participant logs in, they are randomly assigned to a study from a pool of studies that are available.  Participants do not select which study they will complete, and researchers can establish rules for who is eligible to participate in the study.  Participants are unobtrusively selected based on those selection criteria.  For most studies in the pool, once a person has been assigned to the study they are not assigned again in the future.  The Research site is where most new research is conducted at Project Implicit.

PIMH is opening in 2011 and has a format very similar to the Demonstration site.  Participants self-select to participate in tasks about a variety of topics concerning mental health.  Each demonstration study has a few short questionnaires and an implicit measure.  Participants receive feedback on their performance at the end, and can learn about implicit cognition and issues related to mental health with the other materials at the site.

Featured tasks are single studies that might capture particular interest in the public.  In most cases, the featured task is available at the website for a few months.  Sample sizes can be quite large for featured tasks compared to studies conducted at the Research site, but at the cost of participants self-selecting to participate based on the way that the topic is described.

Besides the https://implicit.harvard.edu/ front page, participants engage with the infrastructure through a variety of "private studies."  These are private in the sense that there is a specific URL associated with the study, and researchers usually contact their participants directly to visit that URL.  This method of data collection is used for studies with pre-identified samples, for follow-up studies with samples from the Research site, and for laboratory studies that use the infrastructure for data collection.

Project Implicit from the researcher perspective

Researchers are trained to use the Project Implicit infrastructure with in-person and web-based materials that describe the infrastructure and how to use it.  Researchers do not directly access the production servers and databases that are physically located at Harvard.  Only the project administrators can access those machines.  Instead, researchers get accounts on test environments at the University of Virginia that are replicas of the production environment.  On these servers, researchers can prepare and test their study materials.  Once completed and approved for putting into production, administrators move the files to the production servers.  In some cases, those studies are places into the Research site pool.  In other cases, a private URL is established for the study and researchers are able to direct their participants to that location. 

Researchers use a variety of technologies to create studies.  However, most of it uses simplified templates in the form of text files to define the study materials and procedures.  The actual technical implementation is developed and managed by programmers.  This simplifies the study development process, and significantly shortens the time required for researchers to be able to work with the system. 

Researchers also obtain accounts in the Virtual Lab in order to retrieve data from their own studies or from large datasets that are used by many researchers.  This secure environment allows researchers to make their studies available to co-investigators that have Virtual Lab accounts with a buddy system.   Researchers also use http://peoplescience.org/ as a training and materials resource.