Quantifying egocentric spatial neglect with cancellation tasks: A theoretical validation.

Spatial neglect is characterized by a spatial bias in response to stimuli. The disorder is often assessed with a cancellation task, where several measures can be used to quantify the spatial bias of cancellation responses (e.g., the difference between cancellations on the left and right side, the average location of cancelled targets, and the total number of omissions). Typically, measures of cancellation performance are validated by studying the correlation with measures derived from other tasks to assess neglect (e.g., the directional bisection error derived from performance on the line bisection task). However, the foundation of cancellation performance measures is often more intuitive than theoretical. For instance, it is assumed that measures of cancellation performance isolate the spatial (e.g., the ipsilesional preference typical of spatial neglect) from the non-spatial (e.g., deficits in working memory or sustained attention) sources of error, but this assumption has not been tested yet. Here we formulated a simple model with conceptually meaningful parameters to predict cancellation performance. Our model parameterizes the spatial and non-spatial components of cancellation responses. This model allowed us to study the construct representation of commonly used measures of spatial neglect through the use of Monte Carlo simulations. The results showed that most of the cancellation performance measures are also dependent on non-spatial error sources. The results deepen our understanding of the construct representation of cancellation performance measures, while also having implications for studies focused on the relationship between the spatial and non-spatial attention deficits in spatial neglect.