Slide Slam J3 Sandbox Series
Self-Monitoring During Speech Production in a Picture-Naming and Conflict-Inducing Spatial Stroop Task
Katherine Andrade1, Connor Sperling2, Olivia Toohey3, Carrie McDonald4, David Lee5,6, Leena Kansal5,6, June Yoshii-Contreras5,6, Jerry Shih5,6, Ben-Haim Sharona5,7, Eric Halgren5, Stephanie Ries3; 1Joint Doctoral Program in Language and Communicative Disorders, University of California San Diego & San Diego State University, 2Department of Electrical Engineering, San Diego State University, 3School of Speech, Language, and Hearing Sciences, San Diego State University, 4Department of Psychiatry, University of California San Diego, 5Department of Neuroscience, University of California San Diego, 6Department of Neurology, University of California San Diego, 7Department of Neurosurgery, University of California San Diego
Speakers are continuously monitoring their speech as they are speaking. When an error is detected, they may interrupt themselves and repair their utterance. According to the conflict-based monitoring account (Nozari et al., 2011), self-monitoring during speech production occurs through the detection of conflict between opposing responses subsequently resolved by domain-general cognitive control. Previous neuroimaging studies have identified several regions of the cortex, such as the anterior cingulate cortex (Christoffels et al., 2007) and the superior-temporal gyrus (Tourville et al., 2008), as being associated with speech monitoring. In parallel, electrophysiological studies identified the error negativity (Ne), interpreted as a general-purpose response monitoring mechanism (Ries et al., 2011) associated with, but independent, of error detection. Previously, this component has been shown to be present in both correct and error trials, although larger in errors, and onsetting prior to the onset of articulation in studies using scalp electroencephalography. However, these studies generally lacked combined spatial and temporal resolution allowing to identify which brain region supports different aspects of speech monitoring and conflict resolution. In this study, stereotactic electroencephalographic (SEEG) data was acquired from 19 patients undergoing clinical monitoring for intractable epilepsy. Patients performed various language tasks, including a picture-naming task and a conflict-inducing spatial Stroop paradigm. In the spatial Stroop paradigm, participants were presented with arrows pointing right or left on either side of the fixation cross and had to name the direction of the arrow while ignoring where the arrow was located on the screen. Participants were asked to make verbal responses as quickly and as accurately as possible in both tasks. A total of 17 patients participated in the naming task (5F; mean age 33.7 years, SD 11.4) and 15 patients in spatial Stroop (3F; mean age 36.3 years; SD 11.9). Nine patients (2F; mean age 35.6 years; SD 11.9) produced sufficient speech errors (4 or more true errors) during the picture-naming task (mean error rate = 6.1%, SD = 2.7%). Participants were significantly slower in incongruent than congruent trials in the spatial Stroop paradigm (X^2 (1)= 5.72, p=0.017), there was no significant effect on accuracy rates. Brain activity associated with speech monitoring was investigated by focusing on the Local Field Potential (LFP- 0.1-30Hz) activity. Preliminary results show larger activity in errors than correct trials peaking around vocal onset in bilateral middle-temporal gyrus across multiple subjects during picture-naming. In the spatial Stroop paradigm greater activity is seen peaking before vocal onset for incongruent compared to congruent trials in the left frontal cortex, left medial temporal lobe and peaking around vocal onset in bilateral middle temporal gyrus and left superior temporal gyrus (SEEG analysis performed on 8 patients, 6M; mean age 34.2 years; SD 9.6). These preliminary results suggest a role of the left middle frontal and medial temporal lobe in processes preceding speech monitoring, such as response selection, and a role of the bilateral middle-temporal gyrus and left superior temporal gyrus in conflict resolution for speech monitoring during production.