On the use (and misuse?) of the Edinburgh Handedness Inventory

Researchers who study human cognition and behavior, especially from a neuroscience perspective, often measure subjects' handedness. The most common measure of handedness is the Edinburgh Handedness Inventory (EHI; Oldfield, 1971). Several potential problems with the EHI have been identified during its long history. We informally observed that individual researchers have adopted a variety of modified versions of the EHI, each addressing perceived shortcomings in its own way. To confirm this, we reviewed 899 articles reporting usage of the EHI from 1998 to 2012. For those articles reporting details of the instrument used, we coded information about test items, response format, and scheme for classifying individuals as right-handed. We found tremendous diversity in all three components of the inventory, confirming that many variants of the EHI are used in contemporary research. We furthermore report evidence that researchers who use variants may be unaware that they are not using the original instrument. Variant usage appears to be largely ad hoc and lacking any semblance of uniformity within the scientific community. We discuss how highly variable usage of the EHI may imperil efforts to produce replicable and convergent research findings, and we offer recommendations for future action.