Questions and answers

We offer a multitude of services such as Qualtrics, physical testing spaces, hardware (e.g., EMG, eye-tracking), software (e.g., LifeData, E-prime), and training services to meet your research needs! 

• To learn more about the INSPIRE services such as Qualtrics, software, and training services we offer, please visit Resources. 
• To explore the large selection of physical testing spaces and equipment that INSPIRE offers please visit the Discover the lab page.

• We recommend that you read the guideline papers for the measure(s) of your choice during the design stage and BEFORE data collection: Guidelines Papers - Society for Psychophysiological Research. Do not trust published articles for having implemented good practices! 

• Watch the Mindware videos on data collection (prior to meeting an INSPIRE RA for a live demonstration): MindWare Technologies LTD and X for eye-tracking. 
• Use E-Prime, Experiment Builder or similar to send a signal from the computer for stimuli presentation to the computer for psychophysiological data recording. This reduces risks of experimenter error and potentially problematic time lags. 
• Pilot your study with a small subset of participants (e.g., N = 5) and fully analyze that data to verify all works as it should (e.g., Are the right stimuli presented at the right time? Is randomization working? Are you recording responses and reaction time? Is psychophysiological data clean? How will it be analyzed concretely?). 
• Monitor your data quality during data collection and if possible, preprocess and analyze data on a regular basis to promptly correct issues as they arise.  

1. The most important of these factors is your participant. A few tips here: 

   • Make sure they are comfortable and able to focus on your task (e.g., offer the time to go to the washroom, make sure they are comfortably seated).  

   • Send information ahead of time (e.g., any preparatory tips, including type of clothes to wear, if applicable). 
   • Plan enough time at the beginning of the experiment to build rapport with the participant and allow them to habituate to the experimental context (e.g., around 15 minutes).  
   • Try to find optimal times for scheduling the testing session or instruct the participant to select the optimal time for them. Aim for maximum energy and minimum stress for the participant. 
   • If you notice a lot of noise in your data, stop the experiment and check in with your participant. There is likely a reason that can be quickly addressed - including going to the washroom, drinking a sip of water, scratching that itch, moving a little to reenergize, and simply forgetting to minimize movements. 
   • Plan your experiment to be able to offer breaks regularly to your participants (e.g., every 10 minutes). During the break, participants can move!  
2. Ambient electrical noise (e.g., from lights or computer screens). If you use the INSPIRE facilities, this should already minimized. 
3. Electrode freshness. Disposable electrodes have an expiry date, and they should be used promptly after a bag is opened. If possible, plan your study to use all electrodes in a bag within a relatively short period of time.  
4. Non disposable equipment should be well cared for, including cleaning and minimizing risks of physical damage. 
5. Ultimately, data must also be acquired appropriately. Make sure to have the right acquisition parameters in BioLab (or other), save the parameters and upload these prior to each session.  

INSPIRE staff will ask about your needs at the planning stage. If you foresee needing more material, email us as soon as possible. 

No one and no place is immune to technological failure. The best plan is to act proactively rather than retroactively. Plan enough time to prepare the sessions to ensure that all equipment works as it should (e.g., at least 15 minutes of your scheduled time). Check in with the INSPIRE RA if you encounter any issues. Note that RAs usually start at 9am. Additionally, monitor data quality throughout data collection.  

The short answer is “maybe”. It is possible that you will have to exclude part or all of the data from a participant. If salvaged, the noisy data may reduce power to detect an effect. Best practice is to reduce noise as much as possible during data collection. The rule of thumb is to exclude participants if you need to exclude or estimate more than 10% of the data (but this depends on the population). You may also need a minimum number of trials. A good strategy is to include more trials than necessary. See Q1 for some tips about maximizing data quality.  

This depends on your experiment and requires careful consideration. A few general principles: 

• You may want to add baseline or control tasks throughout the experiment or at the beginning and end. In some cases, the baseline or control task should be close to the period of interest. 

• The baseline may need to have a comparable duration to the task (e.g., if you analyze data by chunks of 5 minutes, possibly have a 5 minutes baseline).  
• The baseline or control task may need to be in the same physical position than the task of interest (seating or standing up).  
• Make sure the participant is relaxed when you record the baseline or control task.

Consult the guideline below. INSPIRE RAs can meet with you too. Forewarned is forearmed:  

• Some researchers are surprised that there is no single good way to preprocess and analyze data, and multiple measures from each technique can be useful. Multiple factors influence the decision process, including characteristics of your data, your research question, and field of research.  

• It is best to learn how to clean data early. Learning how to clean data during the piloting phase will help to: 1) identify problems in your task or data, 2) identify the right parameters for your analyses, 3) properly recognize noise in your data and assess their influence throughout data collection. 
• Make sure to document your preprocessing steps and any information that may help reviewers assess the quality of your data and the rigour of your approach (e.g., % estimated data, % of excluded data and reasons for exclusion, any validity checks like respiratory rate for RSA). 
• The person cleaning the data may need to be blind to group (e.g., clinical group vs. comparison group).