Time Course of Visual Attention to Emotional Stimuli in Persons with Aphasia

Abstract

Event Abstract Back to Event Time Course of Visual Attention to Emotional Stimuli in Persons with Aphasia Sameer A. Ashaie1, 2*, Katie McMenamin2 and Leora Cherney1, 2* 1 Shirley Ryan AbilityLab, United States 2 Feinberg School of Medicine, Northwestern University, United States Depression is a common sequela of aphasia. Cognitive models of depression suggest that visual attention to emotional information plays a key role in the etiology and maintenance of depression (Gotlib et al., 2010). The cognitive underpinnings of depression have been supported by eye-tracking research whereby depressed individuals spent longer time viewing negatively valenced stimuli compared to non-depressed individuals (Armstrong et al., 2012). However, the majority of research has been done by collapsing metrics of visual attention into long time segments (i.e., time segments greater than 1 second) thus obscuring the fine-grained nature of eye-movements (Mirman, 2016). Analyzing attention in smaller time segments may provide a better understanding of how visual attention to emotional stimuli changes over time. We conducted a fine-grained analysis of eye-movements to investigate whether persons with aphasia (due to left-hemisphere stroke) differ in visual attention to emotional stimuli compared to healthy controls. Twenty-four persons with chronic aphasia (post-onset > 6 months) and 13 healthy age and education-matched controls participated. Table 1 summarizes the demographic characteristics. The eye-tracking task was a free-viewing task. Participants were instructed to look at 20 pairs of happy-neutral faces and 20 pairs of sad-neutral faces. The complete presentation time (3500 ms) for each stimulus was analyzed and segmented into 70 consecutive 50ms time bins. Fixation probabilities were computed as a function of emotion trial-type (i.e., happy and sad face) and group (i.e., healthy controls and persons with aphasia). Growth curve analysis was used to investigate the attentional probabilities over time using fourth-order orthogonal time polynomials (linear, quadratic, cubic, and quartic). The four time terms, emotion-trial type and group were entered as fixed effects. Results showed that the face viewing probabilities roughly peaked at 500ms post-stimulus onset and then gradually decreased over time (with a smaller second peak). There was a significant group interaction with quadratic and quartic time terms. We also analyzed emotion bias (emotional face – neutral face) over time. There was a significant interaction between linear, cubic, quartic time terms with group and trial-type. The linear time term interaction suggests that as time progressed, persons with aphasia were less biased to sad faces compared to healthy controls. Both groups demonstrated a strong initial face orienting effect but over time persons with aphasia were less attentive. Persons with aphasia were also biased to neutral faces compared to sad faces early on and then maintained that attention throughout the time course. Overall, findings that time course of eye-movement behavior changes in response to different emotional stimuli suggest that eye-tracking is sensitives to different emotional stimuli in persons with aphasia. Our findings also demonstrate that eye-tracking is feasible in persons with aphasia and there may be a potential to use it as a biomarker for depression. Figure 1 Acknowledgements This research is supported by The Coleman Foundation.