Neural organization of speech production: A lesion-based study of error patterns in connected speech

Brielle C Stark; Alexandra Basilakos; Gregory Hickok; Chris Rorden; Leonardo Bonilha; Julius Fridriksson

We originally identified 114 participants with a naming assessment and 63 with a connected speech assessment; 36 participants had both assessments. To create two mutually exclusive groups with which to compare lesion data, we matched participants with a naming assessment to those with a connected speech assessment using case-control matching in SPSS 25. We matched on aphasia severity (from the Western Aphasia Battery [WAB] Aphasia Quotient [AQ] from Kertesz, 2007) using a FUZZY interval of one standard deviation of the CS task group's AQ. This resulted in two mutually exclusive groups: a connected speech task group of N=63 (CS task group) and a naming task group of N=57 (PNT task group). . CC-BY-NC-ND 4.0 International license is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/544841 doi: bioRxiv preprint Naming (PNT) Task Group (N=57). This task group included those participants who completed a visual confrontation naming task involving the naming of objects, the Philadelphia Naming Test (Roach et al., 1996) , described below in Behavior Tasks. There were 57 participants (32 males; age, M=61.68±12.02 yrs; months post-stroke, M=31.91±37.16) included in this subset with the following aphasia types: anomic, 25;