Induced ideas of reference during social unrest and pandemic in Hong Kong

Abstract

Ideas of reference (IOR) are often implicated in predicting psychosis onset. They have been conceptualized to present on a continuum, from oversensitive psychological reactions to delusional thoughts. It is however unknown to what extent IOR may be triggered by collective environmental stress. We obtained timely data from 9873 individuals to assess IOR in relation to trauma exposure in the 2019–2020 social unrest in Hong Kong. Two levels of IOR are distinguished: attenuated IOR (IOR-A), being the experience of feeling particularly referred to within a group; and exclusive IOR (IOR-E), the experience of feeling exclusively referred to while others are not. Logistic regressions showed that event-based rumination was a shared predictor for IOR-A (OR = 1.07, CI = 1.03–1.10) and IOR-E (OR = 1.09, CI = 1.02–1.17). For IOR-A, three categories of social unrest-related traumatic events (TEs) were significant predictors, including being attacked or having experienced sexual violence (OR = 4.14, CI = 1.93–8.85), being arrested (OR = 4.48, CI = 1.99–10.10), and being verbally abused (OR = 2.66, CI = 1.28–5.53). Being arrested was significant for IOR-E (OR = 3.87, CI = 1.03–14.52), though not when rumination was included. Education level also significantly predicted IOR-E (OR = 0.72, CI = 0.52–0.99). Further analysis revealed that rumination significantly mediated between TEs and IOR severity (β = 0.26, SE = 0.01, CI = 0.24–0.28). The findings are consistent with the hypothesis that IOR-A and IOR-E occur as levels on a continuum, but each has some distinctive correlates. Extrinsic events may play a more prominent role in IOR-A, while intrinsic factors, such as cognitive capacity, may play a more prominent role in IOR-E. The involvement of rumination across the IOR spectrum suggests an opportunity for intervention.

Introduction

Ideas of reference (IOR) can be described as the experience of a spurious sense of relatedness of external events to oneself. IOR are one of the most common groups of symptoms in psychotic disorders, such as schizophrenia (World Health Organization, 1973), and have been observed in a number of other conditions including mood and personality disorders (Colori, 2017; Meyer and Lenzenweger, 2009). Importantly, they also represent prodromal symptoms that may predict psychosis onset (Bendala-Rodríguez et al., 2019). However, little is known concerning whether IOR can be directly triggered by external events, for instance, in the context of large-scale societal conflicts.

Recent studies consider IOR on a psychosis continuum encompassing a range of experiences, from psychologically understandable oversensitivity to self-referential delusions (Rodríguez-Testal et al., 2019; van Os and Reininghaus, 2016). Identifying sets of factors associated with different IOR expressions (corresponding to psychotic-like experiences [PLE], attenuated psychotic symptoms, or psychotic symptoms; Huber and Gross, 1989; Keshavan et al., 2011) may offer insights into how they can emerge in response to intrinsic and extrinsic factors (Bendala-Rodríguez et al., 2019). Past studies have largely viewed IOR as spontaneous experiences linked to individual traits and distant past experiences. While external triggers may play a role in the formation of IOR, their relationships are typically difficult to demonstrate because of the highly individualized and varied contexts associated with spontaneous IOR in everyday life. Shared community stressors may enable investigation into factors associated with IOR induced by well-characterized external events.

Since June 2019, Hong Kong has been undergoing a period of ongoing social unrest. What began as peaceful demonstrations escalated into police-civilian confrontations. Tear gas, rubber bullets, and batons were used for crowd dispersal, while road-blocking and damages to targeted facilities were adopted by some protesters. Over 8000 arrests have been made during this period (Arranz, 2020). Widespread monitoring and surveillance of social media activities have also led to dismissals from jobs and doxxing (for individuals with differing political orientations). During the initial months, substantial levels of mental distress, specifically post-traumatic stress disorder (PTSD) and depressive symptoms, have already been observed (Ni et al., 2020). While societal conflicts remained unresolved, the city was further confronted with waves of coronavirus disease 2019 (COVID-19) outbreak. A recent study has suggested significant increases in PTSD and depressive symptoms as a result of the social unrest, COVID-19, and personal stressful life events (SLEs) in Hong Kong (Wong et al., 2020). Further examination of how these external stressors relate to IOR experiences during ongoing societal conflict and public health crisis would be important.

In studying IOR in a context of societal conflicts, how threat is perceived in an understandable manner may change with events that take place within the community. In this study, we therefore adopted a more stringent criterion for IOR which considers “recipient specificity” as previously described (Wong et al., 2012). We conceptualized recipient specificity as the extent to which an experience is spuriously appraised by the individual as pointing at oneself in particular and to the exclusion of others.

Accordingly, on the milder end of the spectrum, the IOR experiences would involve considering oneself as being particularly referred to more than others, while acknowledging that others are also addressed to a lesser degree, i.e., attenuated IOR (IOR-A). At the other end of the spectrum, a more intense self-referential phenomenon may be less “understandable” and experienced as one being solely and exclusively addressed (i.e., “this [the population-level event] is directed at me and no one else”). The latter of the two, which we term exclusive IOR (IOR-E), would show more resemblance to the aberrant thinking styles encountered in pre-psychotic at-risk mental states (Allswede et al., 2020).

We hypothesized that (1) IOR-A and IOR-E would share some correlates of stressors and vulnerability factors; (2) external social unrest-related traumatic events (TEs) would contribute relatively more to IOR-A and less to IOR-E; (3) rumination, being a transdiagnostic predictor, would be associated with both IOR-A and IOR-E; and (4) individual cognitive functioning, as an intrinsic factor reflected by education level, would be more related to IOR-E.