You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster D48, Thursday, November 9, 6:15 – 7:30 pm, Harborview and Loch Raven Ballrooms

Verb Deficits in Alzheimer’s Disease and Aphasia: Argument-Structure and Thematic-Hierarchy Effects

Caitlyn Antal1, Julie Turbide1, Roberto G. de Almeida1;1Concordia University

Verb knowledge deficits have been reported in individuals with probable Alzheimer’s disease (pAD) and those with aphasia (PWA), in studies manipulating diverse methods and linguistic materials (e.g., Breedin et al., 1998; Grossman et al., 1996; Kemmerer, 2000; Kim & Thompson, 2003, 2004; Manouilidou, et al., 2009). Nevertheless, the underlying nature of verb deficits remains inconclusive. For some, verb deficits have been operationalized in terms of semantic template complexity, with verbs containing hypothetically more internal predicates being more preserved (heavy verbs; e.g., run) than those with less predicates (light verbs; e.g., go) (Breeding et al., 1998). For others, verb deficits arise as a function of argument structure complexity, with verbs requiring a greater number of arguments being more difficult to produce (Thompson, 2003; Bastiaanse & Platonov, 2015). Research on this argument-complexity hypothesis have yielded inconsistent effects. Thompson et al. (2012) reported PWA showing greater production accuracy for one-argument verbs (die) compared to two (destroy) and three (sell) argument verbs, but less difficulty producing three-argument verbs compared to two-argument verbs, an effect which previously found amongst pAD patients (Kim & Thompson, 2004). While some studies have not classified verbs according to semantic or syntactic classes (e.g., Kemmerer & Tranel, 2000), others have relied on classifications that escape linguistic properties. For example, Grossman et al. (1996) classified a verb such as listen as both motion and perception-cognition. In the present study, we employed short movies depicting events and states to probe for three verb classes with different syntactic and semantic properties: perception verbs, lexical causative verbs, and movement verbs. These classes differ in terms of their number of arguments, thematic roles, and hypothetical internal predicates. Lexical causative verbs (peel) were hypothesized to be semantically and syntactically complex, as they project two arguments and may encode multiple internal predicates (x CAUSE y BECOME <peeled>). Perception verbs were hypothesized to be structurally complex but semantically simplex, as they also project two arguments, but encode one predicate (x PERCEIVE y). And movement verbs were hypothesized to be both, semantically and syntactically simplex, with one argument and one predicate (x MOVE). These verbs also differ with regards to thematic-role hierarchy, whereby causative and movement verbs assign the most prominent Agent role to the subject position, and perception verbs assign a less canonical Experiencer role in the subject position. pAD (N = 13), PWA (N = 7), and healthy controls (N = 18) named events and states depicted in short videos clips. Preliminary results revealed that both pAD and PWA show verb deficits in terms of argument structure complexity and thematic role hierarchy. PWA show greater impairments for perception verbs, which are arguably the most difficult for their complex argument structures and non-canonical thematic structures. Similarly, pAD patients showed greater impairments for verbs with both, complex argument structures and thematic roles, namely perception and lexical causative verbs, in comparison to movement verbs. These preliminary results suggest that the underlying principles for the representation of verbs in the brain are argument structures and thematic roles, not internal predicate complexity.

Topic Area: Meaning: Lexical Semantics

Back to Poster Schedule