Adding Context to Browser-Based Physiological Recording
In our earlier Chrome extension demo, we showed how Shimmer3R GSR and PPG data could be streamed while a participant viewed online content.
We have now extended the prototype in three main ways.
First, screenshots can be captured using different triggers, including periodic capture, click-based capture and an optional second screenshot after a configurable delay, allowing time for the view to settle after an interaction such as zooming or navigation.
Second, the recording interface is now a collapsible overlay. Researchers can connect the device, monitor live signals and configure the session, then minimise the overlay so it does not significantly interfere with the participant’s experience.
Third, the export now includes an interactive HTML report. Physiological signals are plotted against event markers, with screenshots and timestamps linked to the relevant points in the session.
Video versus interactive experiences
For a YouTube video, screenshots provide useful confirmation of what was visible, but they are not always essential. Because the extension records the video playback time, researchers can return directly to the corresponding moment in the video and align physiological changes with specific scenes, events or transitions.
Screenshots become more important in non-linear experiences such as a virtual museum tour. Participants may follow different routes, select different exhibits and spend different amounts of time looking at each item. In this setting, click and settled-view screenshots help reconstruct what the participant was actually viewing when changes occurred in the physiological data.
The current demo does not attempt to infer engagement, stress or interest directly. However, it provides a practical way to create richer, context-labelled datasets that could support the development of algorithms that do make those kinds of inferences in the future. By linking physiological signals with media time, interaction events and visual context, the system gives researchers a stronger foundation for building models of engagement, attention, workload or emotional response.
Where this could lead
Verisense has taught us how to manage participants, devices, data collection and processing across research studies.
NeuroLynQ Home demonstrated how participants could wear GSR sensors in their own homes while watching major media events, such as the Super Bowl, and how those individual recordings could be aggregated to understand the broader audience response.
The next natural step is to bring these two capabilities together.
By combining Verisense’s participant and study-management infrastructure with NeuroLynQ’s experience in synchronising, processing and aggregating physiological responses, we can begin building the next stage of NeuroLynQ: a platform for collecting context-rich data from distributed participants and producing aggregated response analytics for media and interactive experiences.
This prototype is an early step towards that direction.
Interested in collaborating, testing the platform or exploring a research use case?
