Loneliness and cognitive function in adults aged 50 years and older: a systematic review; and, Loneliness and psychotic-like experiences in middle-aged and older adults: the mediating role of selective attention to threat and external attribution biases

Abstract

Loneliness is a common experience across the lifespan, including older age. It is linked with multiple adverse health outcomes, such as anxiety, depression, or psychosis at the clinical and subclinical levels. Furthermore, loneliness can influence human cognition and has been associated with poorer cognitive function and biased information processing. With the world's population growing older, it is of increasing importance to understand better how loneliness affects the health and well-being of the ageing population. Furthermore, because of the wide impact on health and the fact that loneliness can be effectively addressed, with the most effective interventions targeting underlying social cognition, loneliness may be a viable target of a broad range of interventions aiming at clinical populations as well as preventative strategies for the general population.

Chapter one of this thesis presents the results of a systematic review of cross-sectional and longitudinal studies on the association between loneliness and cognitive function in middle-aged and older adults (age ≥ 50) without recognised cognitive impairment. A search of six electronic databases supplemented by a manual search yielded twelve articles containing thirteen studies from nine countries. The studies indicated that loneliness was weakly associated with poorer global cognition in older adults without recognised cognitive impairment and predicted accelerated cognitive decline. Most evidence came from studies using the MMSE to assess global cognition. Findings from studies using other measures of global cognition or specific cognitive domains were largely lacking and inconclusive. Furthermore, the limitations of the evidence and the review more broadly are discussed, including the clinical heterogeneity of included studies and measurement bias. Nevertheless, we concluded that interventions promoting healthy cognitive ageing would benefit from addressing loneliness as one of the risk factors. Future research directions are provided.

Chapter two presents the findings of an empirical study testing the hypothesis of whether the association between loneliness and psychotic-like experiences was mediated by two specific cognitive biases: selective attention to threat bias; and a tendency to attribute negative events to external causes, such as other people. A convenience sample (n = 642) of middleaged (40-59 years) and older adults (>60 years) was recruited online. This study found that loneliness was associated with an increased rate of psychotic-like experiences. A parallel mediation model indicated that two cognitive mechanisms mediated this association. Thus, loneliness increased both attention to threat and external attribution bias, which in turn were associated with an increased rate of psychotic-like experiences. This model proved robust after adjusting for possible clinical and socio-demographic confounders, including anxiety and depressive symptoms. The impact of age was observed as moderating factors. This study demonstrates the feasibility of using the psychosis continuum framework in the ageing population as an alternative to clinical studies. The study limitations and directions for future research are also discussed.