GazePlotter
Documentation Chapters

Aggregated Data Export

The Aggregated Data Export pipeline condenses raw eye-tracking vectors into a highly structural, mathematically rigorous long-format CSV. This output is explicitly designed for seamless ingestion into advanced statistical processing engines (R, Python, SPSS).

Analytical Purpose

Generating aggregated data files provides:

  • Explicit statistical metrics calculated individually for each unique Participant-Stimulus-AOI interaction.
  • A standardized long-format CSV architecture optimized for programmatic parsing.
  • Highly parallelized metric outputs consolidated into a singular unified export matrix.

Export Configuration

To generate the analytical export, several systemic parameters must be defined.

Global File Settings

  • File name: Specify the root output filename (the .csv extension is appended automatically).
  • Participant Group: Select which specific logical participant group cohort should be actively iterated during the export generation.

Stimulus Targeting

Designate exactly which stimuli from the core dataset contribute rows to the CSV output.

  • Stimuli Selection: Check the boxes adjacent to the desired stimuli in the Stimuli column. Selecting multiple stimuli combines all resulting permutation data into the same CSV.
  • Delimiter: Choose between Comma (,) or Semicolon (;) to define the column separation character.
  • Decimal Separator: Choose between Dot (.) or Comma (,) for numeric floating-point values.
  • Aggregation Logic: Each individual stimulus interaction resolves into its own discrete rows within the unified matrix.

Metrics Selection

You must deliberately select which specific mathematical operations the system executes.

Duration Metrics

  • Absolute Dwell Time: Total aggregate time logically spent located within the AOI, calculated in milliseconds (ms).
  • Relative Dwell Time (%): Dwell time geometrically expressed as a direct percentage of the total active viewing duration.
  • Time to First Fixation: Total temporal delay from the initiation of the stimulus until the first recorded fixation collides with the AOI. (Outputs -1 if the AOI is never achieved).
  • First Fixation Duration: Absolute duration of the inaugural fixation impacting the AOI. (Outputs -1 if the AOI is never achieved).
  • Mean Fixation Duration: The standard average duration of all independent fixations registered on the AOI. (Outputs -1 if the AOI is never achieved).
  • Mean Visit Duration: The standard average duration of all distinct logical visits to the AOI. (Outputs -1 if the AOI is never achieved).

Frequency Metrics

  • Fixation Count: The absolute integer quantity of discrete fixations registered on the AOI.
  • Visit Count: The absolute integer quantity of grouped continuous visits. (Consecutive, unbroken serial fixations inside the exact same AOI count dynamically as 1 unified visit).

Data Architecture

Understanding the physical structure of the resulting CSV is crucial for programmatic parsing scripts.

CSV Structure (Long Format)

The system automatically outputs a strict long-format structure comprising these specific column headers:

  • Participant_ID: The numeric tracking identifier of the participant.
  • Participant_Name: The semantically defined Participant Display Name.
  • Stimulus: The semantically defined Stimulus Display Name.
  • AOI_Group: The assigned AOI Display Name, or one of the internal system variables.
  • Metric: The string literal representing the specific metric type (e.g., "Dwell_Time").
  • Value: The final integer or floating-point mathematical calculation.

Internal System Variables

Within the AOI_Group column, the exporter may utilize reserved variables:

  • No_AOI: Represents the mathematical aggregation of all fixations that occurred completely outside any manually defined AOI polygons.
  • Any_Fixation: Represents the brute total aggregation calculation across absolutely all recorded fixations for that stimulus iteration.

Structure Example

Participant_ID,Participant_Name,Stimulus,AOI_Group,Metric,Value
1,"P01","Image1","Logo","Absolute_Dwell_Time",1250.5
1,"P01","Image1","Logo","Fixation_Count",3
1,"P01","Image1","No_AOI","Absolute_Dwell_Time",450.2

Execution Workflow

  1. Access Export: Click the Export workspace or data button in the Workspace Toolbar.
  2. Select Format: In the Research Data Formats section, click on the Aggregated Data (CSV) card.
  3. Configure Settings: Define the File name, target Participant Group, and preferred Delimiter and Decimal Separator.
  4. Target Data: Manually select the target Stimuli and target statistical Metrics to be calculated.
  5. Download: Click the Export CSV execution button. The system computes the matrix and serves the CSV. A success toast will display the exact data points and metrics count.

External Tool Optimization: The explicit long format utilized by this export routine was designed from the ground up for data frame ingestion.

  • R: Native compatibility via read.csv() coupled natively with tidyverse restructuring logic.
  • Python: 1:1 mapped importation into pandas dataframes for immediate statistical modeling.
  • SPSS: Immediate direct matrix importation for structural ANOVA processing and regression analysis.