Brain state kinematics and the trajectory of task performance improvement.

Dimensionality reduction techniques offer a unique perspective on brain state dynamics, in which systems-level activity can be tracked through the engagement of a small number of component trajectories. Used in combination with neuroimaging data collected during the performance of cognitive tasks, these approaches can expose the otherwise latent dimensions upon which the brain reconfigures in order to facilitate cognitive performance. Here, we utilized Principal Component Analysis to transform parcellated BOLD timeseries from an fMRI dataset in which 70 human subjects performed an instruction based visuomotor learning task into orthogonal low-dimensional components. We then used Linear Discriminant Analysis to maximise the mean differences between the low-dimensional signatures of fast-and-slow reaction times and early-and-late learners, while also conserving variance present within these groups. The resultant basis set allowed us to describe meaningful differences between these groups and, importantly, to detail the patterns of brain activity which underpin these differences. Our results demonstrate non-linear interactions between three key brain activation maps with convergent trajectories observed at higher task repetitions consistent with optimization. Furthermore, we show subjects with the greatest reaction time improvements have delayed recruitment of left dorsal and lateral prefrontal cortex, as well as deactivation in parts of the occipital lobe and motor cortex, and that the slowest performers have weaker recruitment of somatosensory association cortex and left ventral visual stream, as well as weaker deactivation in the dorsal lateral prefrontal cortex. Overall our results highlight the utility of a kinematic description of brain states, whereby reformatting data into low-dimensional trajectories sensitive to the subtleties of a task can capture non-linear trends in a tractable manner and permit hypothesis generation at the level of brain states.

Real-Life Self-Control is Predicted by Parietal Activity During Preference Decision Making: A Brain Decoding Analysis.

Despite its relevance for health and education, the neurocognitive mechanism of real-life self-control is largely unknown. While recent research revealed a prominent role of the ventromedial prefrontal cortex in the computation of an integrative value signal, the contribution and relevance of other brain regions for real-life self-control remains unclear. To investigate neural correlates of decisions in line with long-term consequences and to assess the potential of brain decoding methods for the individual prediction of real-life self-control, we combined functional magnetic resonance imaging during preference decision making with ecological momentary assessment of daily self-control in a large community sample (N = 266). Decisions in line with long-term consequences were associated with increased activity in bilateral angular gyrus and precuneus, regions involved in different forms of perspective taking, such as imagining one's own future and the perspective of others. Applying multivariate pattern analysis to the same clusters revealed that individual patterns of activity predicted the probability of real-life self-control. Brain activations are discussed in relation to episodic future thinking and mentalizing as potential mechanisms mediating real-life self-control.

A comparison of fMRI and behavioral models for predicting inter-temporal choices.

In an inter-temporal choice (IteCh) task, subjects are offered a smaller amount of money immediately or a larger amount at a later time point. Here, we are using trial-by-trial fMRI data from 363 recording sessions and machine learning in an attempt to build a classifier that would ideally outperform established behavioral model given that it has access to brain activity specific to a single trial. Such methods could allow for future investigations of state-like factors that influence IteCh choices. To investigate this, coefficients of a GLM with one regressor per trial were used as features for a support vector machine (SVM) in combination with a searchlight approach for feature selection and cross-validation. We then compare the results to the performance of four different behavioral models. We found that the behavioral models reached mean accuracies of 90% and above, while the fMRI model only reached 54.84% at the best location in the brain with a spatial distribution similar to the well-known value-tracking network. This low, though significant, accuracy is in line with simulations showing that classifying based on signals with realistic correlations with subjective value produces comparable, low accuracies. These results emphasize the limitations of fMRI recordings from single events to predict human choices, especially when compared to conventional behavioral models. Better performance may be obtained with paradigms that allow the construction of miniblocks to improve the available signal-to-noise ratio.

Predicting Real-Life Self-Control From Brain Activity Encoding the Value of Anticipated Future Outcomes.

Deficient self-control leads to shortsighted decisions and incurs severe personal and societal costs. Although neuroimaging has advanced our understanding of neural mechanisms underlying self-control, the ecological validity of laboratory tasks used to assess self-control remains largely unknown. To increase ecological validity and to test a specific hypothesis about the mechanisms underlying real-life self-control, we combined functional MRI during value-based decision-making with smartphone-based assessment of real-life self-control in a large community sample (N = 194). Results showed that an increased propensity to make shortsighted decisions and commit self-control failures, both in the laboratory task as well as during real-life conflicts, was associated with a reduced modulation of neural value signals in the ventromedial prefrontal cortex in response to anticipated long-term consequences. These results constitute the first evidence that neural mechanisms mediating anticipations of future consequences not only account for self-control in laboratory tasks but also predict real-life self-control, thereby bridging the gap between laboratory research and real-life behavior.

Deterministic response strategies in a trial-and-error learning task.

Trial-and-error learning is a universal strategy for establishing which actions are beneficial or harmful in new environments. However, learning stimulus-response associations solely via trial-and-error is often suboptimal, as in many settings dependencies among stimuli and responses can be exploited to increase learning efficiency. Previous studies have shown that in settings featuring such dependencies, humans typically engage high-level cognitive processes and employ advanced learning strategies to improve their learning efficiency. Here we analyze in detail the initial learning phase of a sample of human subjects (N = 85) performing a trial-and-error learning task with deterministic feedback and hidden stimulus-response dependencies. Using computational modeling, we find that the standard Q-learning model cannot sufficiently explain human learning strategies in this setting. Instead, newly introduced deterministic response models, which are theoretically optimal and transform stimulus sequences unambiguously into response sequences, provide the best explanation for 50.6% of the subjects. Most of the remaining subjects either show a tendency towards generic optimal learning (21.2%) or at least partially exploit stimulus-response dependencies (22.3%), while a few subjects (5.9%) show no clear preference for any of the employed models. After the initial learning phase, asymptotic learning performance during the subsequent practice phase is best explained by the standard Q-learning model. Our results show that human learning strategies in the presented trial-and-error learning task go beyond merely associating stimuli and responses via incremental reinforcement. Specifically during initial learning, high-level cognitive processes support sophisticated learning strategies that increase learning efficiency while keeping memory demands and computational efforts bounded. The good asymptotic fit of the Q-learning model indicates that these cognitive processes are successively replaced by the formation of stimulus-response associations over the course of learning.

Large-scale coupling dynamics of instructed reversal learning.

The ability to rapidly learn from others by instruction is an important characteristic of human cognition. A recent study found that the rapid transfer from initial instructions to fluid behavior is supported by changes of functional connectivity between and within several large-scale brain networks, and particularly by the coupling of the dorsal attention network (DAN) with the cingulo-opercular network (CON). In the present study, we extended this approach to investigate how these brain networks interact when stimulus-response mappings are altered by novel instructions. We hypothesized that residual stimulus-response associations from initial practice might negatively impact the ability to implement novel instructions. Using functional imaging and large-scale connectivity analysis, we found that functional coupling between the CON and DAN was generally at a higher level during initial than reversal learning. Examining the learning-related connectivity dynamics between the CON and DAN in more detail by means of multivariate patterns analyses, we identified a specific subset of connections which showed a particularly high increase in connectivity during initial learning compared to reversal learning. This finding suggests that the CON-DAN connections can be separated into two functionally dissociable yet spatially intertwined subsystems supporting different aspects of short-term task automatization.

Monitor yourself! Deficient error-related brain activity predicts real-life self-control failures.

Despite their immense relevance, the neurocognitive mechanisms underlying real-life self-control failures (SCFs) are insufficiently understood. Whereas previous studies have shown that SCFs were associated with decreased activity in the right inferior frontal gyrus (rIFG; a region involved in cognitive control), here we consider the possibility that the reduced implementation of cognitive control in individuals with low self-control may be due to impaired performance monitoring. Following a brain-as-predictor approach, we combined experience sampling of daily SCFs with functional magnetic resonance imaging (fMRI) in a Stroop task. In our sample of 118 participants, proneness to SCF was reliably predicted by low error-related activation of a performance-monitoring network (comprising anterior mid-cingulate cortex, presupplementary motor area, and anterior insula), low posterror rIFG activation, and reduced posterror slowing. Remarkably, these neural and behavioral measures predicted variability in SCFs beyond what was predicted by self-reported trait self-control. These results suggest that real-life SCFs may result from deficient performance monitoring, leading to reduced recruitment of cognitive control after responses that conflict with superordinate goals.

CREB1 Genotype Modulates Adaptive Reward-Based Decisions in Humans.

Cyclic AMP response element-binding protein (CREB) contributes to adaptation of mesocorticolimbic networks by modulating activity-regulated transcription and plasticity in neurons. Activity or expression changes of CREB in the nucleus accumbens (NAc) and orbital frontal cortex (OFC) interact with behavioral changes during reward-motivated learning. However, these findings from animal models have not been evaluated in humans. We tested whether CREB1 genotypes affect reward-motivated decisions and related brain activation, using BOLD fMRI in 224 young and healthy participants. More specifically, participants needed to adapt their decision to either pursue or resist immediate rewards to optimize the reward outcome. We found significant CREB1 genotype effects on choices to pursue increases of the reward outcome and on BOLD signal in the NAc, OFC, insula cortex, cingulate gyrus, hippocampus, amygdala, and precuneus during these decisions in comparison with those decisions avoiding total reward loss. Our results suggest that CREB1 genotype effects in these regions could contribute to individual differences in reward- and associative memory-based decision-making.

Sparse regularization techniques provide novel insights into outcome integration processes.

By exploiting information that is contained in the spatial arrangement of neural activations, multivariate pattern analysis (MVPA) can detect distributed brain activations which are not accessible by standard univariate analysis. Recent methodological advances in MVPA regularization techniques have made it feasible to produce sparse discriminative whole-brain maps with highly specific patterns. Furthermore, the most recent refinement, the Graph Net, explicitly takes the 3D-structure of fMRI data into account. Here, these advanced classification methods were applied to a large fMRI sample (N=70) in order to gain novel insights into the functional localization of outcome integration processes. While the beneficial effect of differential outcomes is well-studied in trial-and-error learning, outcome integration in the context of instruction-based learning has remained largely unexplored. In order to examine neural processes associated with outcome integration in the context of instruction-based learning, two groups of subjects underwent functional imaging while being presented with either differential or ambiguous outcomes following the execution of varying stimulus-response instructions. While no significant univariate group differences were found in the resulting fMRI dataset, L1-regularized (sparse) classifiers performed significantly above chance and also clearly outperformed the standard L2-regularized (dense) Support Vector Machine on this whole-brain between-subject classification task. Moreover, additional L2-regularization via the Elastic Net and spatial regularization by the Graph Net improved interpretability of discriminative weight maps but were accompanied by reduced classification accuracies. Most importantly, classification based on sparse regularization facilitated the identification of highly specific regions differentially engaged under ambiguous and differential outcome conditions, comprising several prefrontal regions previously associated with probabilistic learning, rule integration and reward processing. Additionally, a detailed post-hoc analysis of these regions revealed that distinct activation dynamics underlay the processing of ambiguous relative to differential outcomes. Together, these results show that L1-regularization can improve classification performance while simultaneously providing highly specific and interpretable discriminative activation patterns.

CACNA1C genotype explains interindividual differences in amygdala volume among patients with schizophrenia.

Affective deficits are one common denominator of schizophrenia (SZ), bipolar disorder (BD) and obsessive compulsive disorder (OCD) with the amygdala indicated as one of the major structures involved in emotion regulation. Previous findings of differences in amygdala volume between healthy controls and patients with SZ, BD or OCD diverge with respect to the affected hemisphere, size and direction of the effect. Variability in the CACNA1C gene has been linked to BD, SZ as well as structural and functional variation in the amygdala in healthy people and patients with BD. We were interested to investigate whether amygdala volumes differ between hemispheres, diagnostic or genotype groups, and whether any interactive effects exist. We combined genotyping of SNP rs1006737 in CACNA1C with structural MRI measurements of relative gray matter (GM) amygdala volume in patients with SZ, BD or OCD as well as healthy controls (N Total = 72). The CACNA1C genotype showed a significant effect on relative GM amygdala volume in patients with SZ. There was a significant left versus right relative GM amygdala volume decrease in patients with SZ or BD. The effects of hemisphere and diagnosis (controls vs. patients with SZ) on relative GM amygdala volume were genotype specific. Our data suggest that the CACNA1C genotype may account for some heterogeneity in the effects of hemisphere and diagnosis on amygdala volume when comparing patients with SZ and controls and point to disturbed Ca(2+)-signaling as a plausible mechanism contributing to the pathology in patients with SZ.

Are electrophysiological correlates of response inhibition linked to impulsivity and compulsivity? A machine-learning analysis of a Go/Nogo task.

Heightened impulsivity and compulsivity are often found in association with both dysfunctional everyday behavior and with psychopathology. Impulsivity and compulsivity are also linked to alterations in behavioral response inhibition and its electrophysiological correlates. However, they are rarely examined jointly and their effect outside of clinical samples is still disputed. This study assesses the influence and interaction of impulsivity and compulsivity as measured by questionnaires (Barratt Impulsiveness Scale, UPPS Impulsive Behavior Scale, and Obsessive-Compulsive Inventory-Revised) on behavioral performance and event-related potentials (N2, P3a, and P3b) in a visual Go/Nogo task. Data from 250 participants from the general population (49% female; age M = 25.16, SD = 5.07) were collected. We used robust linear regression as well as regression tree analyses, a type of machine learning algorithm, to uncover potential non-linear effects. We did not find any significant relationship between the self-report measures and behavioral or neural inhibition effects in either type of analysis, with the exception of a linear effect of the lack of premeditation subscale of the UPPS Impulsive Behavior Scale on behavioral performance. The current sample size was large enough to uncover even small effects. One possibility is that inhibitory performance was unimpaired in a non-clinical sample, suggesting that the effect of these personality traits on inhibition and cognitive control may require a clinical sample or a more difficult task version. Further studies are needed to uncover possible associations and interactions to delineate when impulsivity and compulsivity lead to dysfunctional everyday behavior and psychopathology.

Fast Estimation of L1-Regularized Linear Models in the Mass-Univariate Setting.

In certain modeling approaches, activation analyses of task-based fMRI data can involve a relatively large number of predictors. For example, in the encoding model approach, complex stimuli are represented in a high-dimensional feature space, resulting in design matrices with many predictors. Similarly, single-trial models and finite impulse response models may also encompass a large number of predictors. In settings where only few of those predictors are expected to be informative, a sparse model fit can be obtained via L1-regularization. However, estimating L1-regularized models requires an iterative fitting procedure, which considerably increases computation time compared to estimating unregularized or L2-regularized models, and complicates the application of L1-regularization on whole-brain data and large sample sizes. Here we provide several functions for estimating L1-regularized models that are optimized for the mass-univariate analysis approach. The package includes a parallel implementation of the coordinate descent algorithm for CPU-only systems and two implementations of the alternating direction method of multipliers algorithm requiring a GPU device. While the core algorithms are implemented in C++/CUDA, data input/output and parameter settings can be conveniently handled via Matlab. The CPU-based implementation is highly memory-efficient and provides considerable speed-up compared to the standard implementation not optimized for the mass-univariate approach. Further acceleration can be achieved on systems equipped with a CUDA-enabled GPU. Using the fastest GPU-based implementation, computation time for whole-brain estimates can be reduced from 9 h to 5 min in an exemplary data setting. Overall, the provided package facilitates the use of L1-regularization for fMRI activation analyses and enables an efficient employment of L1-regularization on whole-brain data and large sample sizes.

Disrupted Salience and Cingulo-Opercular Network Connectivity During Impaired Rapid Instructed Task Learning in Schizophrenia.

Rapid instructed task learning (RITL) is the uniquely human ability to transform task information into goal-directed behavior without relying on trial-and-error learning. RITL is a core cognitive process supported by functional brain networks. In patients with schizophrenia, RITL ability is impaired, but the role of functional network connectivity in these RITL deficits is unknown. We investigated task-based connectivity of eight a priori network pairs in participants with schizophrenia (n = 29) and control participants (n = 31) during the performance of an RITL task. Multivariate pattern analysis was used to determine which network connectivity patterns predicted diagnostic group. Of all network pairs, only the connectivity between the cingulo-opercular network (CON) and salience network (SAN) during learning classified patients and control participants with significant accuracy (80%). CON-SAN connectivity during learning was significantly associated with task performance in participants with schizophrenia. These findings suggest that impaired interactions between identification of salient stimuli and maintenance of task goals contributes to RITL deficits in participants with schizophrenia.

Aberrant neural representation of food stimuli in women with acute anorexia nervosa predicts treatment outcome and is improved in weight restored individuals.

Anorexia nervosa (AN) has been associated with altered reward processing. We recently reported greater neural response in secondary visual areas when processing visual food stimuli in acutely underweight AN patients (acAN). In order to examine whether the observed alterations are indicative of acute undernutrition or a potential trait marker of AN, we set out to assess neural responses in acAN and in individuals weight-recovered from AN (recAN). FMRI data were collected from a total of 126 female volunteers, 35 acAN, 33 recAN, and 58 age-matched healthy controls (HC) while they viewed streams of food, social and neutral stimuli. A standard general linear model (GLM) was used to interrogate neural responses to the different stimuli in recAN vs. age-matched HC. Moreover, within-subject multivoxel pattern analyses (MVPA) in the two matched samples (acAN/HC and recAN/HC) were used to estimate neural representation of food vs. neutral, and social vs. neutral stimuli. A multiple regression analysis was conducted to test associations between the accuracy of the neural representation and treatment outcome. The GLM revealed no group differences between recAN and HC. The MVPAs showed greater classification accuracy of food stimuli in the posterior fusiform gyrus in acAN but not recAN. Classification accuracy was associated with better treatment outcome. Our findings suggest that the neural representation of food stimuli is altered in secondary visual areas in acAN and normalizes with weight recovery. Possibly this altered representation reflects attentional engagement motivating food intake, which may promote the recovery process.

Neural representation of newly instructed rule identities during early implementation trials.

By following explicit instructions, humans instantaneously get the hang of tasks they have never performed before. We used a specially calibrated multivariate analysis technique to uncover the elusive representational states during the first few implementations of arbitrary rules such as 'for coffee, press red button' following their first-time instruction. Distributed activity patterns within the ventrolateral prefrontal cortex (VLPFC) indicated the presence of neural representations specific of individual stimulus-response (S-R) rule identities, preferentially for conditions requiring the memorization of instructed S-R rules for correct performance. Identity-specific representations were detectable starting from the first implementation trial and continued to be present across early implementation trials. The increasingly fluent application of novel rule representations was channelled through increasing cooperation between VLPFC and anterior striatum. These findings inform representational theories on how the prefrontal cortex supports behavioral flexibility specifically by enabling the ad-hoc coding of newly instructed individual rule identities during their first-time implementation.

Unbiased Analysis of Item-Specific Multi-Voxel Activation Patterns Across Learning.

Recent work has highlighted that multi-voxel pattern analysis (MVPA) can be severely biased when BOLD response estimation involves systematic imbalance in model regressor correlations. This problem occurs in situations where trial types of interest are temporally dependent and the associated BOLD activity overlaps. For example, in learning paradigms early and late learning stage trials are inherently ordered. It has been shown empirically that MVPAs assessing consecutive learning stages can be substantially biased especially when stages are closely spaced. Here, we propose a simple technique that ensures zero bias in item-specific multi-voxel activation patterns for consecutive learning stages with stage being defined by the incremental number of individual item occurrences. For the simpler problem, when MVPA is computed irrespective of learning stage over all item occurrences within a trial sequence, our results confirm that a sufficiently large, randomly selected subset of all possible trial sequence permutations ensures convergence to zero bias - but only when different trial sequences are generated for different subjects. However, this does not help to solve the harder problem to obtain bias-free results for learning-related activation patterns regarding consecutive learning stages. Randomization over all item occurrences fails to ensure zero bias when the full trial sequence is retrospectively divided into item occurrences confined to early and late learning stages. To ensure bias-free MVPA of consecutive learning stages, trial-sequence randomization needs to be done separately for each consecutive learning stage.

The neural basis of integrating pre- and post-response information for goal-directed actions.

A fundamental prerequisite for goal-directed action is to encode the contingencies between responses (R) producing specific outcomes (O) in specific stimulus conditions (S). The present study aimed to characterize the functional neuroanatomy of different associational sub-components of such S-R-O contingencies during the first few trials of exposure. We devised a novel paradigm that was suited to distinguish BOLD activation patterns related to S-R, R-O, and the full S-R-O contingency. Different from previous studies our experimental design ensured that stimulus-related processes and outcome-related processes were maximally comparable, as both were learned incidentally and lacked intrinsic incentive value, and different from trial-and-error learning situations, outcomes did not serve a special role as performance feedback. We observed contingency-related dissociations between SMA, lateral OFC, and large parts of the reward system including central OFC, anterior striatum and midbrain areas. While the lateral OFC was involved in processing differential outcomes irrespective of a predictive stimulus context, the SMA was specifically engaged when differential outcomes could be predicted by the stimulus. By contrast, the activation pattern of reward system areas suggested that these regions serve a role in integrating non-incentive differential outcome information and incentive common outcome information. Together, these results support the notion that striatal and orbitofrontal regions are involved in outcome-related processes beyond trial-and-error S-R learning, that is, when outcomes are non-incentive and do not serve as reinforcing feedback that drives learning. Furthermore, our results clarify the role of the SMA in outcome-related processes thereby supporting current versions of ideomotor theory.

Integration and segregation of large-scale brain networks during short-term task automatization.

The human brain is organized into large-scale functional networks that can flexibly reconfigure their connectivity patterns, supporting both rapid adaptive control and long-term learning processes. However, it has remained unclear how short-term network dynamics support the rapid transformation of instructions into fluent behaviour. Comparing fMRI data of a learning sample (N=70) with a control sample (N=67), we find that increasingly efficient task processing during short-term practice is associated with a reorganization of large-scale network interactions. Practice-related efficiency gains are facilitated by enhanced coupling between the cingulo-opercular network and the dorsal attention network. Simultaneously, short-term task automatization is accompanied by decreasing activation of the fronto-parietal network, indicating a release of high-level cognitive control, and a segregation of the default mode network from task-related networks. These findings suggest that short-term task automatization is enabled by the brain's ability to rapidly reconfigure its large-scale network organization involving complementary integration and segregation processes.

Investigating the Impact of a Genome-Wide Supported Bipolar Risk Variant of MAD1L1 on the Human Reward System.

Recent genome-wide association studies have identified MAD1L1 (mitotic arrest deficient-like 1) as a susceptibility gene for bipolar disorder and schizophrenia. The minor allele of the single-nucleotide polymorphism (SNP) rs11764590 in MAD1L1 was associated with bipolar disorder. Both diseases, bipolar disorder and schizophrenia, are linked to functional alterations in the reward system. We aimed at investigating possible effects of the MAD1L1 rs11764590 risk allele on reward systems functioning in healthy adults. A large homogenous sample of 224 young (aged 18-31 years) participants was genotyped and underwent functional magnetic resonance imaging (fMRI). All participants performed the 'Desire-Reason Dilemma' paradigm investigating the neural correlates that underlie reward processing and active reward dismissal in favor of a long-term goal. We found significant hypoactivations of the ventral tegmental area (VTA), the bilateral striatum and bilateral frontal and parietal cortices in response to conditioned reward stimuli in the risk allele carriers compared with major allele carriers. In the dilemma situation, functional connectivity between prefrontal brain regions and the ventral striatum was significantly diminished in the risk allele carriers. Healthy risk allele carriers showed a significant deficit of their bottom-up response to conditioned reward stimuli in the bilateral VTA and striatum. Furthermore, functional connectivity between the ventral striatum and prefrontal areas exerting top-down control on the mesolimbic reward system was reduced in this group. Similar alterations in reward processing and disturbances of prefrontal control mechanisms on mesolimbic brain circuits have also been reported in bipolar disorder and schizophrenia. Together, these findings suggest the existence of an intermediate phenotype associated with MAD1L1.