Representative workflow: file intake, quality review, timeline, scanpath and heatmap. The later analysis visuals use demonstration data.
From files to questions
Behaviours Analysis Studio
A Multimodal Data Studio for Human Factors
Before the insight,
understand the recording.
Files arrive with different fields, clocks and gaps. The Studio makes those decisions visible before a visualisation can become a research claim.
THE IDEAMake the path
Make the path
to analysis inspectable.
A workspace for preparing and analysing human-factors data with clear decisions and reproducible records.
The desktop prototype already supports local projects, research-file intake, column mapping and quality review. Further analysis modules are in development. The aim is to make each transformation understandable, from the original source to the eventual result.
THE ENVIRONMENTBring the files in.
Bring the files in.
Keep their meaning visible.
Researchers can create a local project, import source copies, confirm identifiers and map delimited columns with explicit units. Incomplete projects remain usable while unresolved information is reviewed.
Quality scans inspect mapped recordings for data availability and timestamp issues. The timeline, scanpath and heatmap in this website scene demonstrate the intended experience using representative data; they are not analyses of imported research files.
File intake
Bring sources into a project while preserving originals, source checksums and an explicit inventory.
Column mapping
Record which fields mean what, including units and unresolved assumptions.
Quality review
Inspect coverage and timestamp issues before deciding what a recording can support.
RESEARCH VARIABLESWhat makes an
What makes an
analysis inspectable?
These are workflow evaluation questions, rather than a completed experimental design.
THE QUESTION
A column name is not a definition.
A timestamp needs units. A coordinate needs a reference frame. Confirming those meanings should be a visible decision that can be reopened and checked.
- What changes
- Field mappings, units and confirmed metadata
- What we observe
- Unresolved mappings, corrections and recorded mapping decisions
Can another researcher understand how a source field became an analysis input?
From the source
to the decision.
Source identity
Local source copies and checksums establish which files belong to a project.
Preparation record
Saved mappings preserve field conventions and the decisions that need review.
Quality report
Inspectable, exportable reports make data limitations part of the workflow.
The current prototype does not generate normalised analysis tables, automatically synchronise media or calculate research results from imported data. The separate synthetic demonstration illustrates future analysis interactions.
Can you explain
how you got here?
The Studio is being developed around a simple expectation: the route from recording to interpretation should remain open to inspection.
Talk to the lab Next project: Mission Control Interface [OpenMCT]