Metacognitive deficits are common following traumatic brain injury (TBI), and this has important implications for recovery, social relationships, and rehabilitative outcomes (Chiou et al., 2011; Flashman & McAllister, 2002; Ownsworth & Fleming, 2005). Metacognitive deficits have historically been measured using self-report (Allen & Ruff, 1990; Newman et al., 2000; Sherer et al., 1995; Sherer et al., 1998), which is problematic as individuals with an awareness of deficit cannot accurately reflect on their own condition (Akturk & Sahin, 2011). In the past two decades, studies have transitioned to using more objective measures to assess metacognition, including error monitoring tasks (McAvinue et al., 2005; Yeung & Summerfield, 2012) and tasks using retrospective confidence judgments (RCJs) (Busey et al., 2000). Importantly, both tasks are used to study “metacognition,” but clear distinctions as to what domains these tasks measure has not been elucidated. Additionally, both tasks have been linked to executive functioning broadly, but error detection tasks are uniquely associated with measures of attention and self-reported anxiety (Hoerold et al., 2008; O’Keefe et al., 2007), indicating that there may be distinct processes that comprise metacognition. It is a goal to determine what domains these tasks represent so proper assessments of metacognitive ability can be conducted in this population.

Participants included 23 older adults with moderate-severe TBI and 16 age, sex, and education matched healthy control (HC) individuals ages 53-80. All participants received identical neuropsychological test batteries, including two tasks of metacognition (error monitoring task, RCJ task), neurocognitive tasks of attention (Digit Span - Forward, Trail Making Test A) and executive functioning (Digit Span - Backward, Trail Making Test B), and a self-report measure of anxiety (Brief Symptom Inventory - Anxiety subscale). To determine overlapping constructs measured by the two metacognitive tasks, these tasks were correlated with each other and with an attention composite, executive functioning (EF) composite, and anxiety measure in the TBI and HC groups.

In the TBI group, the metacognitive tasks were significantly correlated with each other (r=-0.47, p=0.022). The RCJ task was associated with EF (r=0.47, p=0.025), but not with attention (r=0.20, p=0.358) or anxiety (r=0.25, p=0.248). The error detection task was associated with EF (r=-0.48, p=0.021) and attention (r=-0.46, p=0.026), but not with anxiety (r=-0.19, p=0.383). In the HC group, there were no significant associations between the metacognitive tasks, or between either metacognitive task and EF, attention, or anxiety.

For older adults sustaining TBI, tasks of error detection and tasks using retrospective confidence judgments measured an overlapping construct, with both having an association with executive functioning and only the error detection task being associated with attention. Interestingly, these associations were not found in a healthy control sample of older adults. Both metacognitive tasks have been used in the literature to measure errors of awareness, but this study provides insight that these tasks are measuring different domains of metacognitive ability in older individuals with TBI. Use of multiple tasks of metacognitive ability in this population can help to describe where the deficits of awareness occur following TBI.