Visual Attention to Threat-Related Faces and Delusion-Proneness: An Eye Tracking Study Using Dynamic Stimuli

Abstract

Visual attention to threat-related facial expressions possibly contributes to delusion formation and maintenance and may serve as a vulnerability marker. The aim of the present study was to examine visual attention to threat-related facial expressions using dynamic stimuli in people with differing levels of delusion-proneness. We expected that threat-relevant facial expressions would attract more foveal attention compared to neutral faces. Additionally, we hypothesized that more delusion-prone individuals would show foveal avoidance and that this avoidance would occur particularly in the visual processing of threat-related faces. In a quasi-experimental design we categorized our sample by the paranoia checklist (PCL) into a high-PCL (n = 25) and low-PCL (n = 30) group. The participants' task was to view emotional facial expressions in validated film sequences while eye movements were measured. A mixed ANOVA for the number of fixations and equivalent non-parametric tests for fixation time were conducted. Both groups spent significantly more time viewing relevant features in threat-related faces compared to neutral faces. A significant main effect for group indicated fewer fixations in the high-PCL group compared to the low-PCL group for all faces. A significant Group 9 Affect interaction indicated that the number of fixations differed between the high-PCL group and the low-PCL group depending on whether the faces displayed neutral or threat-related expressions. The findings suggest that higher delusion-proneness is already associated with a deviant style of visually attending to facial expressions even in people who do not have a clinical diagnosis of schizophrenia. This style may thus be a vulnerability marker for psychosis.

Affective prosody and facial emotion recognition in first-episode schizophrenia: Associations with functioning & symptoms

“…One potential factor is abnormalities in visual scan paths found in people with schizophrenia, indicating that these individuals spend less time observing salient features of the face and thus, draw conclusions regarding emotion expression based on a paucity of information (Toh et al, 2011). While this finding has not yet been replicated in first-episode schizophrenia to our knowledge, research indicates that the pattern holds even in participants from the general population who report subthreshold psychotic experiences (Hillmann et al, 2015). In this context, our findings align with cognitive conceptualizations of delusion formation suggesting that paranoid content develops when a person makes judgments based on limited information (Bell et al, 2006; Freeman, 2016) – i.e., assuming our participants have visual scanpath tendencies similar to those found in the literature (biasing visual attention away from salient emotional features), their determination of facial emotion expression based on limited relevant information may be linked to heightened experience of paranoid of persecutory delusions.…”

Selective attention in the news feed: An eye-tracking study on the perception and selection of political news posts on Facebook

“…In the literature, an average deviation of 0.5° is recommended to ensure the best data quality; deviations above 1.0° are seen as unacceptable and should lead to the exclusion of participants (Conklin et al, 2018; Holmqvist et al, 2011). Second, the tracking ratio indicates the robustness of the measurement and describes the proportion to which gaze is recorded in relation to the duration of the entire stimulus expressed in percent (Hillmann et al, 2015). There is no agreement about the minimum limit; however, in previous studies, participants with tracking ratio values below 70–80% were excluded (Duque and Vázquez, 2015; Hillmann et al, 2015).…”

“…Second, the tracking ratio indicates the robustness of the measurement and describes the proportion to which gaze is recorded in relation to the duration of the entire stimulus expressed in percent (Hillmann et al, 2015). There is no agreement about the minimum limit; however, in previous studies, participants with tracking ratio values below 70–80% were excluded (Duque and Vázquez, 2015; Hillmann et al, 2015). Furthermore, we performed a visual inspection of the scanpaths of each participant.…”

Childhood trauma and cognitive biases associated with psychosis: A systematic review and meta-analysis

“…thoughts or speech) as coming from an external source (e.g. media or other people; source monitoring bias [ 9 ]), an over-reliance on prior knowledge over incoming stimuli when interpreting new information (top-down processing bias [ 10 ]), and a bias for interpreting neutral stimuli as threatening (attention to threat bias [ 11 , 12 ]). Finally, biases in inference include hasty decision making in probabilistic inference tasks (the ‘Jumping to Conclusions’ bias; JTC [ 13 ]), and a resistance to revising beliefs in light of new information (a ‘Bias Against Disconfirmatory Evidence’; BADE [ 14 ]) have also been reported in people with psychotic symptoms.…”