The Role of the Angular Gyrus in Goal-directed Behavior-Two Transcranial Magnetic Stimulation Studies Examining Response Outcome Learning and Outcome Anticipation.

Learning the contingencies between a situational context (S), one's own responses (R), and their outcomes (O) and selecting responses according to their anticipated outcomes is the basis of a goal-directed behavior. Previous imaging studies found the angular gyrus (AG) to be correlated to both the representation of R-O associations and outcome-based response selection. Based on this correlational relationship, we investigated the causal link between AG function and goal-directed behavior in offline and online TMS experiments. To this end, we employed an experimental R-O compatibility paradigm testing outcome anticipation during response selection and S-R-O knowledge to probe S-R-O learning. In Experiment 1, we applied 1-Hz rTMS offline to the AG or the vertex before participants performed the experimental tasks. In Experiment 2, we applied online 10-Hz pulse trains to the AG or used sham stimulation during an early action selection stage in half of the trials. In both experiments, the R-O compatibility effect was unaltered when response selection was outcome-based, suggesting no causal role of the AG in outcome anticipation during response selection. However, in both experiments, groups with AG stimulation showed significantly modulated knowledge of S-R-O associations in a posttest. Additionally, in an explorative analysis, we found an induced R-O compatibility effect later in the experiment when response selection was guided by stimulus-response rules, suggesting reduced selectivity of outcome anticipation. We discuss possible compensatory behavioral and brain mechanism as well as specific TMS-related methodical considerations demonstrating important implications for further studies investigating cognitive function by means of TMS.

Changes in global functional network properties predict individual differences in habit formation.

Prior evidence suggests that sensorimotor regions play a crucial role in habit formation. Yet, whether and how their global functional network properties might contribute to a more comprehensive characterization of habit formation still remains unclear. Capitalizing on advances in Elastic Net regression and predictive modeling, we examined whether learning-related functional connectivity alterations distributed across the whole brain could predict individual habit strength. Using the leave-one-subject-out cross-validation strategy, we found that the habit strength score of the novel unseen subjects could be successfully predicted. We further characterized the contribution of both, individual large-scale networks and individual brain regions by calculating their predictive weights. This highlighted the pivotal role of functional connectivity changes involving the sensorimotor network and the cingulo-opercular network in subject-specific habit strength prediction. These results contribute to the understanding the neural basis of human habit formation by demonstrating the importance of global functional network properties especially also for predicting the observable behavioral expression of habits.

Instructing item-specific switch probability: expectations modulate stimulus-action priming.

Both active response execution and passive listening to verbal codes (a form of instruction) in single prime trials lead to item-specific repetition priming effects when stimuli re-occur in single probe trials. This holds for task-specific classification (stimulus-classification, SC priming, e.g., apple-small) and action (stimulus-action, SA priming, e.g., apple-right key press). To address the influence of expectation on item-specific SC and SA associations, we tested if item-specific SC and SA priming effects were modulated by the instructed probability of re-encountering individual SC or SA mappings (25% vs. 75% instructed switch probability). Importantly, the experienced item-specific switch probability was always 50%. In Experiment 1 (N = 78), item-specific SA/SC switch  expectations affected SA, but not SC priming effects exclusively following active response execution. Experiment 2 (N = 40) was designed to emphasize SA priming by only including item-specific SC repetitions. This yielded stronger SA priming for 25% vs. 75% expected switch probability, both following response execution as in Experiment 1 and also following verbally coded SA associations. Together, these results suggest that SA priming effects, that is, the encoding and retrieval of SA associations, is modulated by item-specific switch expectation. Importantly, this expectation effect cannot be explained by item-specific associative learning mechanisms, as stimuli were primed and probed only once and participants experienced item-specific repetitions/switches equally often across stimuli independent of instructed switch probabilities. This corroborates and extends previous results by showing that SA priming effects are modulated by  expectation not only based on experienced item-specific switch probabilities, but also on mere instruction.

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.

Habit strength is predicted by activity dynamics in goal-directed brain systems during training.

Previous neuroscientific research revealed insights into the brain networks supporting goal-directed and habitual behavior, respectively. However, it remains unclear how these contribute to inter-individual differences in habit strength which is relevant for understanding not only normal behavior but also more severe dysregulations between these types of action control, such as in addiction. In the present fMRI study, we trained subjects on approach and avoidance behavior for an extended period of time before testing the habit strength of the acquired stimulus-response associations. We found that stronger habits were associated with a stronger decrease in inferior parietal lobule activity for approach and avoidance behavior and weaker vmPFC activity at the end of training for avoidance behavior, areas associated with the anticipation of outcome identity and value. VmPFC in particular showed markedly different activity dynamics during the training of approach and avoidance behavior. Furthermore, while ongoing training was accompanied by increasing functional connectivity between posterior putamen and premotor cortex, consistent with previous assumptions about the neural basis of increasing habitualization, this was not predictive of later habit strength. Together, our findings suggest that inter-individual differences in habitual behavior are driven by differences in the persistent involvement of brain areas supporting goal-directed behavior during training.

Anticipating the good and the bad: A study on the neural correlates of bivalent emotion anticipation and their malleability via attentional deployment.

In everyday life, we often deliberate about affective outcomes of decisions which can be described as ambivalent; i.e. positive and negative at the same time. For example, when looking forward to meet a dear friend at her/his favorite concert although one dislikes the music that is being performed. Thus, anticipation of bivalent emotions and their volitional regulation is an important ingredient of everyday choices. However, previous studies investigating neural substrates involved in anticipating emotional events mostly focused on anticipating either negative emotions (punishment) or positive emotions (reward) in isolation, thus inducing either of them separately. Furthermore, these studies rather focused on the effortful down-regulation of affect (i.e. reducing negative or positive affect), whereas such conflict situations may also require us to deploy attention on and thereby upregulate anticipated emotions in order to resolve a decision conflict (e.g., by focusing on positive consequences while orienting away from negative consequences of that same situation). To address this gap, we performed a series of three fMRI-experiments using simple visual and auditory stimuli in order to (i) determine the neural correlates involved when anticipating a bivalent affective outcome that is both positive and negative at the same time - related to a conflict situation and (ii) investigate their malleability during anticipation via voluntary emotion regulation using attentional focusing. In these studies, we (i) demonstrate that brain areas involved in anticipating positive (ventral striatum) and negative (anterior insula) emotional events are co-activated when anticipating the occurrence of both punishment and reward at the same time and (ii) provide evidence that attention on either the positive or the negative correlates with a shift in activations of these co-activated neural networks and associated anticipated emotions towards either the positive (increased activity in ventral striatum, ventromedial prefrontal cortex, posterior cingulate cortex) or the negative (increased activity in insula) aspect of the upcoming bivalent outcome. In summary, we provide self-report and neural evidence for the assumption that affective brain systems associated with the processing of bivalent anticipated emotions can be voluntarily controlled by cognitive emotion regulation strategies.

When global rule reversal meets local task switching: The neural mechanisms of coordinated behavioral adaptation to instructed multi-level demand changes.

Cognitive flexibility is essential to cope with changing task demands and often it is necessary to adapt to combined changes in a coordinated manner. The present fMRI study examined how the brain implements such multi-level adaptation processes. Specifically, on a "local," hierarchically lower level, switching between two tasks was required across trials while the rules of each task remained unchanged for blocks of trials. On a "global" level regarding blocks of twelve trials, the task rules could reverse or remain the same. The current task was cued at the start of each trial while the current task rules were instructed before the start of a new block. We found that partly overlapping and partly segregated neural networks play different roles when coping with the combination of global rule reversal and local task switching. The fronto-parietal control network (FPN) supported the encoding of reversed rules at the time of explicit rule instruction. The same regions subsequently supported local task switching processes during actual implementation trials, irrespective of rule reversal condition. By contrast, a cortico-striatal network (CSN) including supplementary motor area and putamen was increasingly engaged across implementation trials and more so for rule reversal than for nonreversal blocks, irrespective of task switching condition. Together, these findings suggest that the brain accomplishes the coordinated adaptation to multi-level demand changes by distributing processing resources either across time (FPN for reversed rule encoding and later for task switching) or across regions (CSN for reversed rule implementation and FPN for concurrent task switching).

Distinct contributions of lateral orbito-frontal cortex, striatum, and fronto-parietal network regions for rule encoding and control of memory-based implementation during instructed reversal learning.

A key element of behavioral flexibility is to quickly learn to modify or reverse previously acquired stimulus-response associations. Such reversal learning (RL) can either be driven by feedback or by explicit instruction, informing either retrospectively or prospectively about the changed response requirements. Neuroimaging studies have thus far exclusively focused either on feedback-driven RL or on instructed initial learning of novel rules. The present study examined the neural basis of instructed RL as compared to instructed initial learning, separately assessing reversal-related instruction-based encoding processes and reversal-related control processes required for implementing reversed rules under competition from the initially learned rules. We found that instructed RL is partly supported by similar regions as feedback-driven RL, including lateral orbitofrontal cortex (lOFC) and anterior dorsal caudate. Encoding-related activation in both regions determined resilience against response competition during subsequent memory-based reversal implementation. Different from feedback-driven RL, instruction-based RL relied heavily on the generic fronto-parietal cognitive control network--not for encoding but for reversal-related control processes during memory-based implementation. These findings are consistent with a model of partly decoupled, yet interacting, systems of (i) symbolic rule representations that are instantaneously updated upon instruction and (ii) pragmatic representations of reward-associated S-R links mediating the enduring competition from initially learned rules.

Towards an understanding of the neural dynamics of intentional learning: Considering the timescale.

Recently, Hampshire et al. (2016) published a paper in NeuroImage investigating the involvement of frontal networks in two types of 'intentional learning'. This included the standard type of deterministic feedback-driven trial-and-error learning and another type of intentional learning that has recently been studied in various facets by means of neuroimaging methods under the terms 'instruction-based learning' (Ruge and Wolfensteller, 2010) or 'rapid instructed task learning' (Cole et al., 2010). By differentiating the learning-related functional roles of different lateral frontal cortex networks and the anterior striatum, Hampshire et al. (2016) contributed valuable results to the field. The aim of this commentary is to increase the interpretability of some of their findings by connecting them to what is already known about fronto-striatal activation dynamics and its functional couplings based on related previous studies. We start with an overview of the rapidly diversifying neuroimaging research on the intentional control of learning and behaviour and its historical embedding. Based thereon we discuss ways to reconcile and integrate the new results presented by Hampshire et al. (2016) particularly regarding the nature of fronto-striatal activation dynamics and their functional couplings during instruction-based learning and during deterministic trial-and-error learning. We conclude that it is important to assess neural activation dynamics on multiple time scales in order to characterize short-term learning and automatization processes as they are evolving across the initial learning trials and further across more extended periods of practice trials.

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.

Distinct fronto-striatal couplings reveal the double-faced nature of response-outcome relations in instruction-based learning.

Higher species commonly learn novel behaviors by evaluating retrospectively whether actions have yielded desirable outcomes. By relying on explicit behavioral instructions, only humans can use an acquisition shortcut that prospectively specifies how to yield intended outcomes under the appropriate stimulus conditions. A recent and largely unexplored hypothesis suggests that striatal areas interact with lateral prefrontal cortex (LPFC) when novel behaviors are learned via explicit instruction, and that regional subspecialization exists for the integration of differential response-outcome contingencies into the current task model. Behaviorally, outcome integration during instruction-based learning has been linked to functionally distinct performance indices. This includes (1) compatibility effects, measured in a postlearning test procedure probing the encoding strength of outcome-response (O-R) associations, and (2) increasing response slowing across learning, putatively indicating active usage of O-R associations for the online control of goal-directed action. In the present fMRI study, we examined correlations between these behavioral indices and the dynamics of fronto-striatal couplings in order to mutually constrain and refine the interpretation of neural and behavioral measures in terms of separable subprocesses during outcome integration. We found that O-R encoding strength correlated with LPFC-putamen coupling, suggesting that the putamen is relevant for the formation of both S-R habits and habit-like O-R associations. By contrast, response slowing as a putative index of active usage of O-R associations correlated with LPFC-caudate coupling. This finding highlights the relevance of the caudate for the online control of goal-directed action also under instruction-based learning conditions, and in turn clarifies the functional relevance of the behavioral slowing effect.

The dynamics of functional brain network segregation in feedback-driven learning.

Prior evidence suggests that increasingly efficient task performance in human learning is associated with large scale brain network dynamics. However, the specific nature of this general relationship has remained unclear. Here, we characterize performance improvement during feedback-driven stimulus-response (S-R) learning by learning rate as well as S-R habit strength and test whether and how these two behavioral measures are associated with a functional brain state transition from a more integrated to a more segregated brain state across learning. Capitalizing on two separate fMRI studies using similar but not identical experimental designs, we demonstrate for both studies that a higher learning rate is associated with a more rapid brain network segregation. By contrast, S-R habit strength is not reliably related to changes in brain network segregation. Overall, our current study results highlight the utility of dynamic functional brain state analysis. From a broader perspective taking into account previous study results, our findings align with a framework that conceptualizes brain network segregation as a general feature of processing efficiency not only in feedback-driven learning as in the present study but also in other types of learning and in other task domains.

Functional integration processes underlying the instruction-based learning of novel goal-directed behaviors.

How does the human brain translate symbolic instructions into overt behavior? Previous studies suggested that this process relies on a rapid control transition from the lateral prefrontal cortex (LPFC) to the anterior striatum (aSTR) and premotor cortex (PMC). The present fMRI study investigated whether the transfer from symbolic to pragmatic stimulus-response (S-R) rules relies on changes in the functional coupling among these and other areas and to which extent action goal representations might get integrated within this symbolic-pragmatic transfer. Goal integration processes were examined by manipulating the contingency between actions and differential outcomes (i.e. action goals). We observed a rapid strengthening of the functional coupling between the LPFC and the basal ganglia (aSTR and putamen) and orbitofrontal cortex (OFC) as well as between the LPFC and the anterior dorsal PMC (pre-PMd), the anterior inferior parietal lobule (aIPL), and the posterior superior parietal lobule (pSPL). Importantly, only some of these functional integration processes were sensitive to the outcome contingency manipulation, including LPFC couplings with aSTR, OFC, aIPL, and pre-PMd. This suggests that the symbolic-pragmatic rule transfer is governed by principles of both, instrumental learning (increasingly tighter coupling between LPFC and aSTR/OFC) and ideomotor learning (increasingly tighter coupling between LPFC and aIPL/pre-PMd). By contrast, increased functional coupling between LPFC and putamen was insensitive to outcome contingency possibly indicating an early stage of habit formation under instructed learning conditions.

(Still) longing for food: insulin reactivity modulates response to food pictures.

Overweight and obesity pose serious challenges to public health and are promoted by our food-rich environment. We used functional magnetic resonance imaging (fMRI) to investigate reactivity to food cues after overnight fasting and following a standardized caloric intake (i.e., a 75 g oral glucose tolerance test, OGTT) in 26 participants (body mass index, BMI between 18.5 and 24.9 kg m(-2)). They viewed pictures of palatable food and low-level control stimuli in a block design and rated their current appetite after each block. Compared to control pictures, food pictures activated a large bilateral network typically involved in homeostatically and hedonically motivated food processing. Glucose ingestion was followed by decreased activation in the basal ganglia and paralimbic regions and increased activation in parietal and occipital regions. Plasma level increases in insulin correlated with cue-induced appetite at the neural and behavioral level. High insulin increases were associated with reduced activation in various bilateral regions including the fusiform gyrus, the superior temporal gyrus, the medial frontal gyrus, and the limbic system in the right hemisphere. In addition, they were accompanied by lower subjective appetite ratings following food pictures and modulated the neural response associated with it (e.g., in the fusiform gyrus). We conclude that individual insulin reactivity is critical to reduce food-cue responsivity after an initial energy intake and thereby may help to counteract overeating.

Focusing on Future Consequences Enhances Self-Controlled Dietary Choices.

Self-controlled dietary decisions, i.e., choosing a healthier food over a tastier one, are a major challenge for many people. Despite the potential profound consequences of frequent poor choices, maintaining a healthy diet proves challenging. This raises the question of how to facilitate self-controlled food decisions to promote healthier choices. The present study compared the influence of implicit and explicit information on food choices and their underlying decision processes. Participants watched two video clips as an implicit manipulation to induce different mindsets. Instructions to focus on either the short-term or long-term consequences of choices served as an explicit manipulation. Participants performed a binary food choice task, including foods with different health and taste values. The choice was made using a computer mouse, whose trajectories we used to calculate the influence of the food properties. Instruction to focus on long-term consequences compared to short-term consequences increased the number of healthy choices, reduced response times for healthy decisions, and increased the influence of health aspects during the decision-making process. The effect of video manipulation showed greater variability. While focusing on long-term consequences facilitated healthy food choices and reduced the underlying decision conflict, the current mindset appeared to have a minor influence.

Strategy-effects in prefrontal cortex during learning of higher-order S-R rules.

All of us regularly face situations that require the integration of the available information at hand with the established rules that guide behavior in order to generate the most appropriate action. But where individuals differ from one another is most certainly in terms of the different strategies that are adopted during this process. A previous study revealed differential brain activation patterns for the implementation of well established higher-order stimulus-response (S-R) rules depending on inter-individual strategy differences (Wolfensteller and von Cramon, 2010). This raises the question of how these strategies evolve or which neurocognitive mechanisms underlie these inter-individual strategy differences. Using functional magnetic resonance imaging (fMRI), the present study revealed striking strategy-effects across regions of the lateral prefrontal cortex during the implementation of higher-order S-R rules at an early stage of learning. The left rostrolateral prefrontal cortex displayed a quantitative strategy-effect, such that activation during rule integration based on a mismatch was related to the degree to which participants continued to rely on rule integration. A quantitative strategy ceiling effect was observed for the left inferior frontal junction area. Conversely, the right inferior frontal gyrus displayed a qualitative strategy-effect such that participants who at a later point relied on an item-based strategy showed stronger activations in this region compared to those who continued with the rule integration strategy. Together, the present findings suggest that a certain amount of rule integration is mandatory when participants start to learn higher-order rules. The more efficient item-based strategy that evolves later appears to initially require the recruitment of additional cognitive resources in order to shield the currently relevant S-R association from interfering information.

Rapid formation of pragmatic rule representations in the human brain during instruction-based learning.

The present functional magnetic resonance imaging study investigated the instruction-based learning of novel arbitrary stimulus-response mappings in order to understand the brain mechanisms that enable successful behavioral rule implementation in the absence of trial-and-error learning. We developed a novel task design that allowed the examination of rapidly evolving brain activation dynamics starting from an explicit instruction phase and further across a short behavioral practice phase. As a first key result, the study revealed that different sets of brain regions displayed either decreasing or increasing activation profiles already across the first few practice trials, suggesting an impressively rapid redistribution of labor throughout the brain. Furthermore, behavioral performance improvement across practice was tightly coupled with brain activation during the practice phase (caudate nucleus), the instruction phase (lateral midprefrontal cortex), or both (lateral premotor cortex bordering prefrontal cortex). Together, the present results provide first important insights into the brain systems involved in the rapid transfer of control from initially abstract rule representations induced by explicit instructions toward pragmatic representations enabling the fluent behavioral implementation.

Costly habitual avoidance is reduced by concurrent goal-directed approach in a modified devaluation paradigm.

Avoidance habits potentially contribute to maintaining maladaptive, costly avoidance behaviors that persist in the absence of threat. However, experimental evidence about costly habitual avoidance is scarce. In two experiments, we tested whether extensively trained avoidance impairs the subsequent goal-directed approach of rewards. Healthy participants were extensively trained to avoid an aversive outcome by performing simple responses to distinct full-screen color stimuli. After the subsequent devaluation of the aversive outcome, participants received monetary rewards for correct responses to neutral object pictures, which were presented on top of the same full-screen colors. These approach responses were either compatible or incompatible with habitual avoidance responses. Notably, the full-screen colors were not relevant to inform approach responses. In Experiment 1, participants were not instructed about post-devaluation stimulus-response-reward contingencies. Accuracy was lower in habit-incompatible than in habit-compatible trials, indicating costly avoidance, whereas reaction times did not differ. In Experiment 2, contingencies were explicitly instructed. Accuracy differences disappeared, but reaction times were slower in habit-incompatible than in habit-compatible trials, indicating low-cost habitual avoidance tendencies. These findings suggest a small but consistent impact of habitual avoidance tendencies on subsequent goal-directed approach. Costly habitual responding could, however, be inhibited when competing goal-directed approach was easily realizable.

Low-Frequency TMS Results in Condition-Related Dynamic Activation Changes of Stimulated and Contralateral Inferior Parietal Lobule.

Non-invasive brain stimulation is a promising approach to study the causal relationship between brain function and behavior. However, it is difficult to interpret behavioral null results as dynamic brain network changes have the potential to prevent stimulation from affecting behavior, ultimately compensating for the stimulation. The present study investigated local and remote changes in brain activity via functional magnetic resonance imaging (fMRI) after offline disruption of the inferior parietal lobule (IPL) or the vertex in human participants via 1 Hz repetitive transcranial magnetic stimulation (rTMS). Since the IPL acts as a multimodal hub of several networks, we implemented two experimental conditions in order to robustly engage task-positive networks, such as the fronto-parietal control network (on-task condition) and the default mode network (off-task condition). The condition-dependent neural after-effects following rTMS applied to the IPL were dynamic in affecting post-rTMS BOLD activity depending on the exact time-window. More specifically, we found that 1 Hz rTMS applied to the right IPL led to a delayed activity increase in both, the stimulated and the contralateral IPL, as well as in other brain regions of a task-positive network. This was markedly more pronounced in the on-task condition suggesting a condition-related delayed upregulation. Thus together, our results revealed a dynamic compensatory reorganization including upregulation and intra-network compensation which may explain mixed findings after low-frequency offline TMS.