Task-related functional connectivity of the caudate mediates the association between trait mindfulness and implicit learning in older adults

Abstract

Accumulating evidence shows a positive relationship between mindfulness and explicit cognitive functioning, i.e., that which occurs with conscious intent and awareness. However, recent evidence suggests that there may be a negative relationship between mindfulness and implicit types of learning, or those that occur without conscious awareness or intent. Here we examined the neural mechanisms underlying the recently reported negative relationship between dispositional mindfulness and implicit probabilistic sequence learning in both younger and older adults. We tested the hypothesis that the relationship is mediated by communication, or functional connectivity, of brain regions once traditionally considered to be central to dissociable learning systems: the caudate, medial temporal lobe (MTL), and prefrontal cortex (PFC). We first replicated the negative relationship between mindfulness and implicit learning in a sample of healthy older adults (60–90 years old) who completed three event-related runs of an implicit sequence learning task. Then, using a seed-based connectivity approach, we identified task-related connectivity associated with individual differences in both learning and mindfulness. The main finding was that caudate-MTL connectivity (bilaterally) was positively correlated with learning and negatively correlated with mindfulness. Further, the strength of task-related connectivity between these regions mediated the negative relationship between mindfulness and learning. This pattern of results was limited to the older adults. Thus, at least in healthy older adults, the functional communication between two interactive learning-relevant systems can account for the relationship between mindfulness and implicit probabilistic sequence learning.

Age group differences in learning-related activity reflect task stage, not learning stage

“…Of note, no participant was able to accurately identify relationships between the cues and targets, supporting the notion that the associations were learned without awareness and were not the result of different explicit learning strategies between younger and older adults. Although the learning and age effects were largely consistent with previous studies using the TLT (Howard et al, 2013(Howard et al, , 2008Seaman et al, 2013;Simon et al, 2012;Stillman et al, 2016aStillman et al, , 2016b and other IAL tasks (Aizenstein et al, 2006;Bennett et al, 2007;Curran, 1997;Fera et al, 2005;Howard et al, 2004), we did not find that the age group difference in the triplet type effect was larger during late learning, as others have (Howard et al, 2008;Simon et al, 2012). This is likely due to methodological differences with the current TLT version that may have facilitated learning in both age groups (e.g., manipulating joint but not conditional probability, using fewer unique triplets to equate the number of HF and LF trials; see Franco et al, 2020) or differences in the definition of early versus late stages from previous studies (e.g., number of exposures to each triplet, number of unique triplets, length of presentation; Rieckmann et al, 2010;Simon et al, 2012).…”

“…The current study extends prior work by examining the inter-region correlations as a function of both learning stage and age group, as well as assessing the mediating role of prefrontal cortex. Earlier studies have reported similar positive relationships between IALrelated activity in the medial temporal cortex (extending beyond hippocampus) and basal ganglia (caudate) in younger and older adults using functional connectivity analyses of fMRI data acquired during TLT performance (Stillman et al, 2016b) and between the hippocampus and basal ganglia (caudate, putamen, nucleus accumbens) in younger adults using task-related fMRI during other associative learning tasks Stark, 2015, 2011). Converging results across these paradigms provides confidence that the interactions among these regions are reflective of associative learning processes rather than minor task idiosyncrasies.…”

“…IAL declines in older relative to younger adults, with larger age group differences emerging in later stages of learning when the sequences are more well-learned (Bennett et al, 2007;Curran, 1997;Howard et al, 2013Howard et al, , 2008Howard et al, , 2004Seaman et al, 2013;Simon et al, 2012;Stillman et al, 2016aStillman et al, , 2016b. This finding may reflect differential engagement of the neural substrates that mediate late versus early stages of IAL in younger and older adults or age-related differences in the extent to which different brain regions coordinate learning of associations at each stage.…”

“…Based on these models, one might predict a cooperative (positive) or competitive (negative) relationship between hippocampus and basal ganglia activity (Freedberg et al, 2020), particularly during early IAL when they are both engaged in learning associations. Most studies that have correlated IAL-related activity in these regions collapsed across all learning stages (Dennis and Cabeza, 2011;Poldrack et al, 2001;Rieckmann et al, 2010;Seger and Cincotta, 2005;Stillman et al, 2016b), but two studies found cooperative relationships during the peak of learning that decoupled over the course of learning Stark, 2015, 2011) and another showed that these hippocampus-basal ganglia interactions were mediated by activity in the prefrontal cortex across learning stages (Poldrack and Rodriguez, 2004). However, the extent to which these inter-region relationships reflect coordination of associative processing during IAL may be constrained by anatomical connections (i.e., corticohippocampal and corticostriatal loops).…”

White matter microstructural correlates of associative learning in the oldest-old

“…In the laboratory, behavioral evidence of implicit associative learning (IAL) is seen as faster and/or more accurate responses to stimuli that can be predicted based on their relationship to prior events, such as frequently occurring cue-cue-target associations in the Triplet Learning Task (TLT; Howard et al, 2008). IAL has been established in adults across the lifespan, albeit with smaller learning effects in younger-old (65-85 years) relative to young (20-30 years) adults (Bennett et al, 2007;Curran, 1997;Howard et al, 2013Howard et al, , 2008Howard et al, , 2004Seaman et al, 2013;Simon et al, 2012;Stillman et al, 2016aStillman et al, , 2016c. But IAL has not yet been demonstrated in oldest-old adults (90+ years) in spite of extensive characterization of their impairments across other cognitive domains (Melikyan et al, 2019).…”