“…For our study, we selected a set of 30 commands for common interactions from the current literature, which were divided into four groups, along with their referents, which were presented with a stimulus before and after the command ( Fig. 12): a group of 4 basic commands (''Turn light on/off'' and ''Brighten/dim light'') [23], [47], a group of 13 selection commands (e.g., ''Select a date in digital/analog format'') [48], a group of 9 commands that are based on sets (e.g., ''Expand/collapse a set of menu items'', ''Select an item from the full set of data'', and ''Select (again) from the selected subset'') [48], [49], and a group of 4 referents for specification (''Specify a position, a direction, a translation, and a rotation'') [50]. These tasks have been chosen to cover an increasing number of dimensions, from 0D (e.g., ''Turn light on''), 1D (e.g., ''Dim light''), and 2D (e.g., ''Specify a position''), to 2D1/2 (e.g., ''Specify a translation'').…”