Comparing Eye-tracking Data of Children with High-functioning ASD, Comorbid ADHD, and of a Control Watching Social Videos

The eye-tracking method can help researchers in the field of special educational needs to explore the differences in visual attention performances for children with and without autism spectrum disorders, ASD, when they view videos of social scenarios. The main advantage of the eye-tracking technique is that the eye-gaze paths and vision fixation durations on various relevant social stimuli can be captured, quantified, and compared between children with and without autism. Visual demonstration of this method is critical.

As the steps on choosing the target area of interest in the video for analysis, may be difficult to illustrate purely in written format. Begin, by escorting the child into the eye-tracking room and having them sit in front of the 23-inch color LCD monitor with the eye-tracker 60 centimeters away. Sit next to the child, and instruct them to look at the screen.

Have another research investigator operate the eye-tracker from the other side of the partition. Have the child watch the calibration dots that set the viewing boundaries across the screen as it captures the eye movements using infrared corneal reflectance technology. After calibration, instruct the participants to view the social videos one after another, while eye movement data is captured during viewing.

Provide them with three social scenario videos to watch in total. Once the experiment is complete, analyze the data by first choosing context-relevant targets, such as the face or hands in their initial 500 milliseconds of appearance in each scene of the video as areas of interest, or AOIs. Label the AOIs in the information box on the left panel.

Upon the completion of the addition and selection of the AOIs in the current frame, move the cursors in the timeline bar at the bottom panel to the next frame. Then, manually adjust the location and boundary of the AOIs in each frame of the video in the presentation video software of the eye-tracker, as the target areas change each timeframe of the video due to the movement of the people or objects as the story of the social video develops. If some existing AOIs are present for 500 milliseconds in the current scene, or if they are not relevant in the new frame in the video, right-click these AOIs to deactivate them in the new frame.

Then, to run a statistical analysis of the eye-tracking indices, first choose the media file for analysis. Click Analyze the Selected Media. Choose the descriptive statistics, and select the dependent measures in Metrics.

Then, choose Recordings in Rows, and select AOI Media Summary in Columns. Pick Update to analyze the eye-tracking patterns. Lastly, create the scan path of a scene from the data by choosing Visualization and Gaze Plot in the software.

Select the media and recordings in the left panel for visualization. In the bottom timeline, move the lower cursor to the beginning of the target scene, and move the upper cursor to the end of the target scene. Ensure Accumulate is chosen for the data field to show the accumulative scan path.

Then, finally click Export and Visualization Image to save the scan path as a separate image file. Here, the colored ovals are the areas of interest. For example, the face, eyes, mouth, hands, mobile phone, and the bag of the lady, which show the first moments in one of the scenes in Video One.

For a social scene, the blue dots trace the scan paths for the neuro-typical control child, and green dots for the ASD child, and the red dots for the ASD-ADHD child. The dots in the figure indicate the locations of visual fixations. The first fixation duration was longer for the neuro-typical child than for the ASD and ASD-ADHD children.

The total fixation duration was shorter for the ASD-ADHD child, than for both the neuro-typical child and the ASD child. Further, the total number of fixation counts was the largest for the ASD child, second for the neuro-typical child, and the shortest for the ASD-ADHD child. While eye-tracking evidence confirms that children with ASD have difficulty in perceiving multiple social stimuli simultaneously and contextually, whereas children with comorbid ASD-ADHD tend to miss important momentary information, which leads them to misjudge social situations.