Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Linguistic and non-linguistic time comprehension after stroke
Poster Session A - Sandbox Series, Thursday, October 24, 10:00 - 11:30 am, Great Hall 3 and 4
This poster is part of the Sandbox Series.
Nicoletta Biondo1,2, Eleonora Ghiraldini3, Vera J. Wheatley2, Alexis L. Pracar2, Maria V. Ivanova2, Simona Mancini1, Nina F. Dronkers2,4; 1BCBL, Spain, 2UC Berkeley, USA, 3Florida State University, USA, 4UC Davis, USA
People with aphasia often produce incorrectly tensed verbs, and their time comprehension abilities, albeit less investigated, can also be impaired (Clahsen & Ali, 2009; Faroqi-Shah & Dickey, 2009; Jonkers & de Bruin, 2009; Fyndanis et al., 2013). Interestingly, the source of time impairment seems to be related to time reference and the ability to access discourse (Bastiaanse et al., 2011; Bos et al., 2014), rather than morphological complexity (Faroqi-Shah & Thompson, 2007). Discourse operates by rules that go beyond the level of the sentence (Avrutin, 2006; Druks, 2017). Therefore, an open question is whether this time impairment is purely linguistic or more domain-general. In this project, we are carrying out a systematic investigation of time processing in post-stroke survivors by addressing two questions: (i) which aspects of time comprehension are impaired? (ii) Is this impairment purely linguistic? The project currently involves English speakers with left hemisphere lesions, ≥ 6 months post-stroke, and adequate hearing/vision/motor skills to complete the tasks, as well as controls. To address (i), we designed two tasks to test Event Location (e.g., “The boy peeled the banana”, task: select one of the three dots representing past/present/future on a timeline) and Temporal Concord (e.g., “Tomorrow the boy peeled the banana”, task: acceptability judgment with two buttons: good or bad) in the same pool of participants. To address (ii), we designed a non-linguistic version of the Event Location task, where the events were represented through colored line drawings rather than sentences. Preliminary results (N = 13 stroke survivors; N = 34 controls) of logit mixed-effect model analyses showed that: (i) Both Event Location and Temporal Concord abilities are significantly impaired in stroke survivors compared to controls (Event Location: z=-4.44, p<.0001; Temporal Concord: z=-4.16, p<.0001). Interestingly, locating events in the past is harder than locating events in the present (z=-4.1, p<.0001), while locating events in the future is only numerically but not significantly harder than locating events in the present (z=-1.79, p=.07), both for stroke survivors and controls. Conversely, judging the temporal concord between a verb and a preceding mis/matching adverb did not significantly change as a function of the past/future tense of the verb (z=-0.76, p=.45). These results are currently consistent with accounts suggesting that past is more difficult to process compared to present and future (Past Discourse Linking Hypothesis, Bastiaanse et al., 2013). (ii) Compared to the control group, stroke survivors struggle in locating events on a timeline, even when temporal information is conveyed non-linguistically (z=-2.18, p<.05). Moreover, locating an event outside of the present, either in the past (z=-3.59, p<.001) or in the future (z=-2.30, p<.05) is more difficult in the non-linguistic task, for both groups. These preliminary findings suggest that the time impairment may not be solely language-specific. The recruitment of Spanish stroke survivors in the second phase of the project will help us tease apart whether the past-future dissociation found in the linguistic and non-linguistic data is due to linguistic strategies related to the different realization of past/future verb morphology in English.
Topic Areas: Disorders: Acquired, Syntax and Combinatorial Semantics