Working Memory Updating and Maintenance in Obesity

In order to navigate goal-directed behavior in an ever-changing complex environment, a fine balance between maintaining and updating of goal-relevant information is required. A cognitive system crucially involved in these processes is our working memory (WM). While maintenance of WM representations is thought to be implemented by recurrent excitatory networks in the PFC (Seamans et al., 2003, Curtis & D’Esposito, 2003; Fallon et al., 2017), updating is accomplished via fronto-striatal go/no-go pathway modulations (O’Reilly and Frank, 2006, Fallon & Cools, 2014; Fallon, van der Schaaf, ter Huurne, & Cools, 2017). Importantly, neurocomputational models, and fMRI and PET studies have demonstrated the involvement of dopaminergic signaling in the modulation of these two complementary processes (Durstewitz & Seamans, 2008; Hazy, Frank, & O’Reilly, 2007; O’Reilly & Frank, 2006). In line with this, flexible maintenance and updating of WM representation has been associated with certain single-nucleotide polymorphisms (COMT and DRD2), both of which are indicative for frontal and striatal dopamine levels, respectively. (Käenmäki et al., 2010; Syvänen et al., 1997; Hirvonen et al., 2005; Leyton et al., 2004; Montgomery et al., 2003). With respect to the COMT gene, Met allele carriers (who supposingly have high prefrontal DA concentrations) are reported to perform better on tasks that require stable maintenance of working memory representations (Berryhill et al., 2013, Farrell et al., 2012; Savitz, Solms, & Ramesar, 2006). Val carriers (who supposingly have low prefrontal DA concentrations) however, perform better on tasks that mainly probe cognitive flexibility (Cameron et al., 2018; Colzato et al., 2010). With respect to the DRD2/ANKK1-TaqIA polymorphism it has been shown that carriers of the A1 allele have up to 30% reduced D2 receptor density in striatum and outperform non-carriers on measures of cognitive flexibility (Jönsson et al., 1999; Pohjalainen et al., 1998; Stelzel et al., 2009). Importantly, recent studies suggest that these two SNPs act on WM interactively, where there are balanced and unbalanced combinations of genotypes (Gracia-Gracia et al., 2011; Reuter, Schmitz, Corr, & Hennig, 2006). In obese individuals, the accurate adaptation of WM representations seems to be compromised. Behaviorally, obese participants have been shown to perform worse on task that require cognitive adaptation (Steenbergen & Colzato, 2017). Furthermore, obese individuals prefer short-term rewards despite of negative long-term outcome (Horstmann et al., 2011; Mathar et al., 2015) or repeat previously rewarded action, despite of current devaluation (Horstmann et al., 2015b). These findings hence suggest that obese individuals might fail to update the negative consequences of their actions, which could contribute to dysfunctional preservation of maladaptive (eating-) behaviors. Additionally, functional changes in the above mentioned WM-related brain areas have been observed in obesity. Several studies report altered central dopamine levels and dopamine receptor density in striatum (Horstmann, Fenske, and Hankir 2015a, Wang et al., 2001; Dunn et al., 2012; Eisenstein et al., 2013; Kessler et al., 2014; Cosgrove et al., 2015; Horstmann et al., 2015a), as well as changes in grey matter density in PFC in obese individuals (Horstmann et al., 2011; Mathar et al., 2015). While there is general consensus that obesity is associated with alterations in WM functioning (Yang, et al., 2018, Gonzales et al., 2010; Coppin et al., 2014), it remains unclear to what extent and which of the above mentioned WM mechanisms contribute to these impairments. In the present study, we therefore aim to tackle this gap and elucidate on potential impairments in the specific sub processes of WM functioning in obesity. To this end, we implement a task-based fMRI, using a modified version of a delayed match-to-sample task originally designed by Fallon and Cools (2014), which compiles specific conditions that probe updating and maintenance of mental representations. Participants have to remember two target stimuli, signaled by the letter “T” centered between the two stimuli, and match with a probe presented at the end of each trial. The task consists of three task conditions: update, ignore, and maintain. In the update condition participants are first presented with two target stimuli (signaled by the centered “T”). Then they get to see a new pair of target stimuli that displaces the previous target stimuli and has to be matched with the subsequently presented probe. In the ignore condition participants are presented a new set of non-target stimuli (signaled by a centered “N”) after the first set of target stimuli. They have to ignore the non-target stimuli and match the previously presented target stimuli with the subsequent probe. In the maintain condition no distractor stimuli are presented, participants only see one pair of target stimuli that they have to match with the subsequent probe. Instead of the non-target stimuli (maintain condition with matched target-probe delay to the ignore condition) or the first target stimuli (maintain condition with matched target-probe delay to the update condition) a fixation cross is presented. Stimuli and fixation cross remain on the screen for 2000 msec in the first phase of a trial, and 1000 msec in the second phase of a trial. The time of the probe being presented and thus the time for the participant to make a choice is 2000 msec. Feedback whether the choice made at the end of each trial (“Is the present probe one of the two target stimuli?” Yes or No) is not provided. The task is separated in four blocks, with feedback on performance between each block. The three task conditions were equally distributed over all four blocks. Each block consists of 32 trials, amounting to a total of 128 trials. Each trial is separated by a inter-trial interval with a jitter of m 2000 to 6000 msec. The task stimuli are unique, randomly computer-generated, monochromatic RGB “spirographs”. Outcome measures are accuracy and reaction time. The total duration of the task is approximately 30 minutes. At the time of submission, data acquisition has started. All datasets have been collected. The data has not been looked at yet, except for general hypotheses-independent sanity checks (e.g. motion and segmentation checks in the fMRI data, performance above chance checks for the task data)