“…A vast literature has found that VWM capacity limits of about 3-5 independent objects are (a) robust over many different types of objects and response methods, (b) minimally affected by the number of dimensions of each object, (c) selectable from larger sets of displayed objects, and (d) potentially composed of meaningful patterns, words, or "chunks" (e.g., Averbach & Coriell, 1961;Eriksen & Lappin, 1967;Fougnie & Marois, 2006;Lappin, 1967;Lappin & Ellis, 1970;Luck & Vogel, 1997Miller, 1956;Pratte, Park, Rademaker, & Tong, 2017;Sperling, 1960;Vogel, Woodman, & Luck, 2001;Woodworth & Schlossberg, 1954;Zhang & Luck, 2011). An important question, however, is whether this limitation involves only memory for previous observations.…”