A single experiment does not necessarily fit into a single Inquisit script. In many cases it may be necessary to collect data using a combination of different Inquisit scripts. Ideally,the different scripts and programs would run as a single, seamless sequence of tasks requiring no manual intervention from the researcher. This article discusses approaches to achieve this.
The Inquisit language provides a simple facility for stringing together multiple scripts into a single data collection session, the batch element. The batch element allows you to specify a list of script files to run in order.
<batch> / file="bart.iqx" / file="iowagamblingtask.iqx" / file="cardsort.iqx" </batch>
The batch element must be defined in its own seperate script, called a "batch script". Batch scripts can be run the same way other scripts are run.
Batch scripts enable more than just running a battery of tests. It's also possible to define different sets of scripts to be run for different groups of participants. To accomplish this, the batch script would include a separte batch element for each condition or group within the study. These batch elements can then be filtered by the group numbers assigned to participants so that a particular batch element is selected and run for each group number. Consider the following example:
<batch> / groups = (1 of 2) / file="treatment.iqx" / file="digitspan.iqx" / file="cardsort.iqx" </batch> <batch> / groups = (2 of 2) / file="control.iqx" / file="digitspan.iqx" / file="cardsort.iqx" </batch>
The batch script contains two batch elements corresponding to a treatment and control condition. The /groups attribute determines the group to which a given batch element applies. The first batch element is run for participants with group number 1, and the second element for participants with group number 2. Participants can be pre-assigned to a group number, or they can be assigned randomly on the fly.
In longitudinal studies it's common or participants to run different sets of tests at different phases of the study. For this kind of study design, batch elements can be defined for different sessions, similarly to how they can be defined for different groups. Instead of filtering based on group number, however, the batch elements are filtered based on the session number. The following example shows a longitudinal study with 3 data collection sessions:
<batch> / sessions = (1 of 3) / file="alcoholdotprobe.iqx" / file="alcoholiat.iqx" / file="alcoholstroop.iqx" </batch> <batch> / sessions = (2 of 3) / file="alcoholiat.iqx" </batch> <batch> / sessions = (3 of 3) / file="alcoholdotprobe.iqx" / file="alcoholiat.iqx" / file="alcoholstroop.iqx" / file="anglingrisktask.iqx" </batch>
On the first session, the participant completes the Alcohol Dot Probe, IAT, and Stroop. On the second session, only the Alcohol IAT is run. On the third session, all three of the tests are run in addition to the Angling Risk Task. Inquisit Web tracks the session number for each participant automatically so that the first time a participant runs the experiment, their session number is 1, the second time is 2, etc. Session numbers can be manually entered using Inquisit Lab.