Negative affective traits and the discrepancy between perceived and logged smartphone behavior: A response surface analysis

A burgeoning literature has linked the use of new technologies to an increasing burden of mental health problems. However, one important methodological concern with this association has been the confounding effect that poor mental health can have on self-reported estimates of technology use. Because the existing literature has focused on self-reported use to measure technology use, this bias may lead to inappropriate inferences being made about the impact of technology on its users. To examine this further, Response Surface Analysis was applied to data from 511 smartphone users to disentangle the effects of logged and self-reported smartphone use on depression, in a conceptual replication of Sewall and Parry (2021). This analysis is then extended to include negative urgency, a facet of impulsiveness associated with increased risky decision-making when experiencing strong negative emotions. The results broadly replicate those of Sewall and Parry (2021), by finding a rising ridge asymmetric congruence effect, driven by higher depression scores amongst respondents who overestimate their smartphone usage. Depression scores were higher when smartphone estimates were congruent and high, and among people who overestimated their actual use. Effects of negative urgency were weaker than depression and inconsistent. These findings add further support to the hypothesis that measurement bias confounds the association between smartphone use and mental health.