Francesco Vespignani

What allows for an immediate interpretation of negated meaning? 

Classical models of negated statements processing assume that for a sentence like ‘the door is not closed’ the negated meaning is derived in two stages: first the negated state of affairs (the door is closed) is accessed in terms of abstract (Clark & Chase, 1972) or embodied (e.g. Ludke et al. , 2008) non-strictly linguistic representations and in a second stage the actual meaning of the utterance (an open door) is derived. The processing of negation is thus typically assumed to be a relevant exception of the idea that utterance meaning is always extracted by an immediate and fully incremental automatic process. Relevant but limited empirical data against this hypothesis have been offered. Nieuwland and Kuperberg (2008) assume immediate access to actual meaning is derived by empirically defined pragmatic felicity, Tian, Ferguson, and Breheny (2010) assume specific pragmatic devices (negative QUD) are relevant, and Oreans et al. (2014) focus on contextually sensitive predictability of the alternatives (binary vs. multuary). Through three sentence-picture matching behavioral studies, we found an advantage in accessing negative meaning when negation is used to assert nonexistence (the circle is not colored) with respect to opposition (the circle is not black), even considering task-specific predictability. The data are informative about the issue of negative meaning access, allowing one to distinguish between principled pragmatic factors and task-related general domain factors. Moreover, the results show an interplay between accuracy and reaction times, which, in our view, require the development of models that go beyond the classical debate between one and two stages.