top of page
Picture of 11 year old boy holding up two pieces of paper.  In one he shows a triangle that he was asked to color in prior to treatment.  In the other, a semi-circle he was asked to color in only a day after treatment.  The contrast in the output of the two coloring activities shows how much vision improves right away, and how it allows a child access to a seemingly simple activity, even without any visual experience.  The child can see the contrast in his performance and seems extremely happy and proud to show off his improvement.

RA, treated at age 11, proudly shows off his work: a triangle he colored in prior to treatment, and semi-circle he colored in a day after treatment shows the significant improvement in his vision for a simple coloring activity that every child should be able to do.

Drawing in Visual Impairments

Drawing is a powerful tool for examining and fostering skill acquisition in children, across various developmental and perceptual differences. However, the challenges arise when addressing learning and sensory differences, particularly in accessing universally designed graphic content essential for STEM learning. This challenge stems from a lack of comprehension regarding how 2D graphical cues are represented across the spectrum of typical and atypical development, creating obstacles for educators and engineers who lack evidence to optimize graphic content. My work involves quantifying the impact of typical and atypical visual experiences on the ability to draw and perform other visual constructional skills. It advocates for a reevaluation of graphical material delivered to children with different encoding strategies.

Watch this short talk I gave at the Vision Science Society Annual Meeting, 2022

Key Findings

1

Reconstruction of 2D patterns following treatment for blindness

Roughly defined as Visuospatial Construction, the ability to organize individual elements into a whole relies on visuo-spatial processing, motor planning, and executive functioning. Despite numerous assessments characterizing these skills, the fundamental contribution of early visual experience for acquiring them remains unclear. In this study, we examine how children who gain vision late in life learn a simple 2D reconstruction task. We find that, despite minimal visual experience, the onset of vision allows immediate reconstruction of a generally coherent 2D pattern. However, the absence of early visual experience leads to persistent spatial distortion errors and a continued reliance on localized pairwise error correction as a primary strategy. This suggests that children with early visual deprivation are unable to integrate individual elements into a whole. These errors may be permanent, regardless of later visual experience, resulting in pattern deformation and mapping errors that the children seem unaware of, despite their rich visual experiences since treatment.

Poster of this ongoing work, presented by the talented Grace Hu at VSS 2023

Image of a poster presented at the Vision Science Society annual meeting in May 2023 of our preliminary results on the 2D pattern reconstruction project described above.

2

Copying Visually and Haptically Presented Shapes after Late Sight-Onset

In this study, we tease apart the impact of different sensory experiences for copying simple patterns, considering visual, haptic, and motor experiences, as well as their integration. Our systematic assessment of copying performance involves children with typical and atypical visual development across three tasks: tracing a visual pattern, copying a visually presented pattern, and copying a haptically presented pattern. We discover that vision onset without visual experience suffices for tasks that rely on local matching and do not require encoding of the full shape (e.g., tracing). However, tasks requiring global pattern representation and motor planning (e.g., copying) demand extensive visual experience, regardless of the input pattern's nature (visual or haptic). Intriguingly, even with prior extensive haptic experience, blind children necessitate time to gain visual experience to create a visual copy of a haptically explored pattern; haptic experience has no facilitatory effect for this task, likely because visual and haptic encoding are not interchangeable when matching to a visual-motor plan for production.

Watch poster presentation at Psychonomics Society Annual Meeting, 2022:

See poster presented at Cognitive Neuroscience Society Annual Meeting, 2020:

Poster of preliminary findings on the shape copying task described above.  This was presented at the Cognitive Neuroscience Society Annual Meeting in 2020.  For a more updated version of these data, refer to the poster video presentation above.

3

Drawing objects from memory following treatment for blindness

In contrast to copying, drawing a known object requires both knowledge about the object and a rich representation capturing its appearance from various viewpoints. Research on typical development reveals a consistent, multi-stage sequence in drawing emergence across children and cultures. Studies with early blind individuals highlight the central role of visual experience, particularly for later stages of drawing development, like object-centered representations.

 

To explore the timing of visual experience, we employed behavioral and computational methods, assessing the drawing performance of children born blind who gained vision later in life. Despite significant improvements in motor skills, visuomotor integration, and visual memory, some aspects of their drawing development remain limited even after acquiring sight and extensive visual experience. While late visual experience enhances recognizability and object-diagnostic detail, certain stages of drawing development, never emerge spontaneously, and seem to be dependent on visual experience occurring early in development.  For example, despite a year of vision, children continue to produce a floor plan depiction of a house, never capturing what a house would like like from a particular point of view.  Similarly, they show relative positional errors of components and orientation errors.  Their errors suggest that in the absence of early visual experience with the 3D world, children fail to build an object-centered representation that typically emerges through experiencing many different view-centered descriptions that are registered on the retina. 

See poster presented at MIT Cognitive Science Meetup, 2021:

4

Influence of persistent visual feedback for visual-motor skill acquisition

In this study, we investigated the role of visual feedback in enhancing fine visuomotor skills, even in late stages of development. We compared performance improvements of adult participants on a tracing task with the non-dominant hand using  three training regimens: 'Persistent' visual-feedback (seeing the trace emerge in real-time), 'Non-Persistent' visual-feedback (seeing the final trace but not the emerging process), and no training. Our finding that 'Persistent' visual feedback leads to significantly better tracing performance than either of the other two training protocols highlights the feasibility of rapid visuomotor learning in adulthood, and emphasizes the impact of visual veridicality and the crucial role of dynamically emergent visual information in driving this learning.

This study has led to a working hypothesis regarding the role of dynamically emergent visual and haptic cues in facilitating graphic recognition and production in children with sensory impairments. Currently, this hypothesis is undergoing testing in the context of enhancing accessibility to graphic material for children with visual impairments.

Read our paper:

Unell, A., Eisenstat, Z. M., Braun, A., Gandhi, A., Gilad-Gutnick, S.*, Ben-Ami, S. and Sinha, P. (2021).  Influence of visual feedback persistence on visuo-motor skill improvement.  Scientific Reports, 11: 17347.

bottom of page