Background We are considering the statistical analysis of functional magnetic resonance imaging (fMRI) data. As demonstrated in previous work, grouping voxels into regions (of interest) and carrying out a multiple test for signal detection on the basis of these regions typically leads to a higher sensitivity when compared with voxel-wise multiple testing approaches. Methods In the case of a multi-subject study, we propose to define the regions for each subject separately based on their individual brain anatomy, represented, e.g., by regional labels. The aggregation of the subject-specific evidence for the presence of signals in the different regions is then performed by means of a combination function for p-values. We validate the proposed methodology with simulated data and apply it to real fMRI data of a hypothesis-driven approach towards identifying brain regions involved in understanding software code. Results The results of our simulated data indicate that the proposed approach improves power relative to voxel-wise inference and performs comparably to a cluster-based baseline. Testing our method on real fMRI data, we found that our approach yields overlapping results with a two-stage approach for which two independent experiments are needed, one for defining the regions and one for actual signal detection. Conclusions In this paper, we overall demonstrate that our method of utilizing anatomical information is a candidate to provide a more sensitive analysis of fMRI data.

Background We are considering the statistical analysis of functional magnetic resonance imaging (fMRI) data. As demonstrated in previous work, grouping voxels into regions (of interest) and carrying out a multiple test for signal detection on the basis of these regions typically leads to a higher sensitivity when compared with voxel-wise multiple testing approaches. Methods In the case of a multi-subject study, we propose to define the regions for each subject separately based on their individual brain anatomy, represented, e.g., by regional labels. The aggregation of the subject-specific evidence for the presence of signals in the different regions is then performed by means of a combination function for p -values. We validate the proposed methodology with simulated data and apply it to real fMRI data of a hypothesis-driven approach towards identifying brain regions involved in understanding software code. Results The results of our simulated data show that our proposed method demonstrated significantly higher power in detecting true activation. Testing our method on real fMRI data, we found that our approach yields overlapping results with a two-stage approach for which two independent experiments are needed, one for defining the regions and one for actual signal detection. Conclusions In this paper, we overall demonstrate that our method of utilizing anatomical information is a candidate to provide a more sensitive analysis of fMRI data.

The hemodynamic response (HR) in event-related functional magnetic resonance imaging is typically assumed to be stationary. While there are some approaches in the literature to model nonstationary HRs, few focus on rapid changes. In this work, we propose two procedures to investigate rapid changes in the HR. Both procedures make inference on the existence of rapid changes for multi-subject data. We allow the change point locations to vary between subjects, conditions and brain regions. The first procedure utilizes available information about the change point locations to compare multiple shape parameters of the HR over time. In the second procedure, the change point locations are determined for each subject separately. To account for the estimation of the change point locations, we propose the notion of post selection variance. The power of the proposed procedures is assessed in simulation studies. We apply the procedure for pre-specified change point locations to data from a category learning experiment.

Learning is an essential skill that humans and animals need to interact successfully with stimuli in their surrounding environment. Discovering the learning motifs applied by humans and animals is a highly complex task. Indeed, identifying significant motifs in behavioural actions by analyzing how actions are affected by stimuli and prior actions enables us to understand how a learning motif is formed by a subject. Granger causality (GC) is a statistical tool used to check whether the past of a variable X is predictive of the future of another variable Y, in this case, we say that X Granger causes Y. Furthermore, subjects may also change the learning motif followed over time due to the learning process. In this study, we propose a method that (1) models a ‘learning motif’ as a set of Granger causality relationships involving past stimuli and past actions, (2) employs a hidden Markov model (HMM) to capture the change in the followed learning motifs and (3) identifies the salient learning motifs that are common to many subjects. The evaluation of our proposed method is not a trivial task due to the absence of the ground truth of the learning motifs. In general, it is difficult to acquire the ground truth of the learning motifs because for example, animals cannot articulate these learning motifs and this also applies to humans in complex learning tasks. Therefore, we also propose a solution that does not require ground truth to validate the derived learning motifs. We evaluate our proposed method on behavioural data collected from two groups of mice, a group of healthy mice and a group of mice with induced cognitive impairment. We show that our model is appropriate for identifying the learning motifs followed by these two groups of animals during a learning experiment in which animals are expected to maximize the reward gain.

Many challenges in life come without explicit instructions. Instead, humans need to test, select, and adapt their behavioral responses based on feedback from the environment. While reward-centric accounts of feedback processing primarily stress the reinforcing aspect of positive feedback, feedback's central function from an information-processing perspective is to offer an opportunity to correct errors, thus putting a greater emphasis on the informational content of negative feedback. Independent of its potential rewarding value, the informational value of performance feedback has recently been suggested to be neurophysiologically encoded in the dorsal portion of the posterior cingulate cortex (dPCC). To further test this association, we investigated multidimensional categorization and reversal learning by comparing negative and positive feedback in an event-related functional magnetic resonance imaging experiment. Negative feedback, compared with positive feedback, increased activation in the dPCC as well as in brain regions typically involved in error processing. Only in the dPCC, subarea d23, this effect was significantly enhanced in relearning, where negative feedback signaled the need to shift away from a previously established response policy. Together with previous findings, this result contributes to a more fine-grained functional parcellation of PCC subregions and supports the dPCC's involvement in the adaptation to behaviorally relevant information from the environment.

The lateralization of processing in the auditory cortex for different acoustic parameters differs depending on stimuli and tasks. Thus, processing complex auditory stimuli requires an efficient hemispheric interaction. Anatomical connectivity decreases with aging and consequently affects the functional interaction between the left and right auditory cortex and lateralization of auditory processing. Here we studied with magnetic resonance imaging the effect of aging on the lateralization of processing and hemispheric interaction during two tasks utilizing the contralateral noise procedure. Categorization of tones according to their direction of frequency modulations (FM) is known to be processed mainly in the right auditory cortex. Sequential comparison of the same tones according to their FM direction strongly involves additionally the left auditory cortex and therefore a stronger hemispheric interaction than the categorization task. The results showed that older adults more strongly recruit the auditory cortex especially during the comparison task that requires stronger hemispheric interaction. This was the case although the task difficulty was adapted to achieve similar performance as the younger adults. Additionally, functional connectivity from auditory cortex to other brain areas was stronger in older than younger adults especially during the comparison task. Diffusion tensor imaging data showed a reduction in fractional anisotropy and an increase in mean diffusivity in the corpus callosum of older adults compared to younger adults. These changes indicate a reduction of anatomical interhemispheric connections in older adults that makes larger processing capacity necessary when tasks require functional hemispheric interaction.

Software is created by people who think, feel, and express themselves to one another and their computers. For a long time, researchers have investigated how people read and write code on their computers and talk about code with one another. This way, researchers identified skills, education, and practices necessary to acquire expertise and perform software development duties. While these investigations are valuable, we have yet to devise and validate a scientific theory of program comprehension , which would be an important step in designing support for developers that is tailored to their cognitive needs. To succeed, we need techniques to shed more light on how programmers think. To this end, we need to look beyond computer science research. Specifically, in the field of psychology and cognitive neuroscience, considerable progress has been made in building theories of cognitive processes. Important enabling technologies include eye tracking, functional magnetic resonance imaging (fMRI), electroencephalography (EEG), and functional near infrared spectroscopy (fNIRS). These methods have revolutionized the understanding of cognitive processes and are routinely used in non-computing disciplines. Such techniques have the potential to also modernize classic approaches to program comprehension research by informing new experimental designs. However, the use of such technologies to study program comprehension is recent, and many of the challenges of this interdisciplinary field remain unexplored. This report documents the program and the outcomes of Dagstuhl Seminar 22402, “Foundations for a New Perspective of Understanding Programming”, which explores these challenges. In total, 23 on-site participants attended the seminar along with two virtual keynote speakers. Participants engaged in intensive collaboration, including discussing past and current research, identifying gaps in the literature

Instantaneous Granger causality has been used in economy and physiological systems as a measure for quantifying directed effects from prior and contemporary observations to observations in the future. However, standard approaches are mostly unable to capture the instantaneous and the Granger causality together in non-stationary categorical time series where the exact order in which past observations occur has no influence on the causality. In this paper, we propose a novel machine learning-based instantaneous Granger causality (IGC) method that summarizes the past of the cause time series ignoring the temporal order of past observations within a window, and quantifies the dependency between these summaries and the effect time series at each time point independently. Furthermore, our approach allows monitoring the evolution of IGC over time. We apply our method on behavioral data collected from 76 participants in an auditory category learning experiment. The learning process can be seen as an evolution in the 'decision policies' across trials, where the current policy is derived from prior stimuli and (accumulated feedback to) prior responses of the participant. The dependency of the current response on the prior exposure to the stimuli and the subsequent responses makes IGC a useful tool in analyzing how the stimulus features contribute to the decision policy: we demonstrate that the instantaneous Granger causalities between stimulus features and responses can distinguish learners from non-learners at early phases of the experiment. Our evaluation shows that our new method outperforms the typically used approach 'Instantaneous Transfer Entropy' (ITE).

Dynamic Time Warping (DTW) is a method generally used to align pairs of time series with different lengths, which is for instance applied in speech recognition. In this study, we use a category learning experiment (CLE) as use-case, in which the participants have to learn a specific target from a pool of predefined categories within a certain amount of time. From a companion system-based point of view, it is important to detect certain anomalies related to affective states, such as surprise or frustration, elicited during the course of learning. In this work, we analyse the button press dynamics (BPD) data from an auditory CLE with the goal of detecting anomalies of the aforementioned type. To this end, we first select a small set of participants, for which we have definite ground truth labels. Subsequently, we apply DTW in combination with hierarchical clustering to separate the anomaly-specific data from the remaining samples. We compare the outcomes to clustering results based on the extraction of intuitive task-specific features. Our results indicate that applying the DTW approach in combination with Single Linkage Clustering in order to detect CLE-related anomalies is preferable to its feature extraction-based alternative, in person-independent scenarios.

The analysis of button press dynamics (BPD) may reveal more detailed information about decisions in human–computer interaction. One huge advantage of a BPD-specific analysis lies in its tangible nature. More precisely, each button press (BP) constitutes a two-dimensional signal consisting of time and intensity coordinates, which can be easily illustrated. Moreover, one can also interpret BP signals by extracting several intuitive features, such as the duration and the maximum intensity. In this study, we analyse the characteristics of such intuitive BPD features. To this end, we conduct cluster analysis experiments with the following evaluation protocol. First, for each person-specific set of BPs, we will define ground truth (GT) clusterings by evaluating the BPs as time series and applying dynamic time warping (DTW) for the calculation of distances between two instances. Subsequently, we will extract a small set of intuitive features. Finally, we will compute the similarity between the DTW- and feature-based clusterings, based on two popular similarity measures. The outcomes of our experiments lead to the following observation. Extending binary BP information by one additional feature, i.e. the maximum press intensity, can already significantly improve the analysis of BP evaluations.

Distractibility is one of the key features of attention deficit hyperactivity disorder (ADHD) and has been associated with alterations in the neural orienting and alerting networks. Task-irrelevant stimuli are thus expected to have detrimental effects on the performance of patients with ADHD. However, task-irrelevant presentation of novel sounds seems to have the opposite effect and improve subsequent attentional performance particularly in patients with ADHD. Here, we aimed to understand the neural modulations of the attention networks underlying these improvements. Fifty boys (25 with ADHD) participated in a functional magnetic resonance imaging (fMRI) study in which unique (novel) or repeatedly presented (familiar) sounds were placed before a visual flanker task in 2/3 of the trials. We found that presenting any sound improved task performance in all participants, but the underlying neural mechanisms differed for the type of sound. Familiar sounds led to a stronger increase in activity in the left posterior insula in patients with ADHD compared to typically developing peers. Novel sounds led to activations of the fronto-temporoparietal ventral attention network, likewise in ADHD and TD. These changes in signaling by novelty in the right inferior frontal gyrus were directly related to improved response speed showing that neural orienting network activity following novel sounds facilitated subsequent attentional performance. This mechanism of behavioral enhancement by short distractions could potentially be useful for cognitive trainings or homework situations.

Auditory event-related fields (ERFs) measured with magnetoencephalography (MEG) are useful for studying the neuronal underpinnings of auditory cognition in human cortex. They have a highly subject-specific morphology, albeit certain characteristic deflections (e.g., P1m, N1m, and P2m) can be identified in most subjects. Here, we explore the reason for this subject-specificity through a combination of MEG measurements and computational modeling of auditory cortex. We test whether ERF subject-specificity can predominantly be explained in terms of each subject having an individual cortical gross anatomy, which modulates the MEG signal, or whether individual cortical dynamics is also at play. To our knowledge, this is the first time that tools to address this question are being presented. The effects of anatomical and dynamical variation on the MEG signal is simulated in a model describing the core-belt-parabelt structure of the auditory cortex, and with the dynamics based on the leaky-integrator neuron model. The experimental and simulated ERFs are characterized in terms of the N1m amplitude, latency, and width. Also, we examine the waveform grand-averaged across subjects, and the standard deviation of this grand average. The results show that the intersubject variability of the ERF arises out of both the anatomy and the dynamics of auditory cortex being specific to each subject. Moreover, our results suggest that the latency variation of the N1m is largely related to subject-specific dynamics. The findings are discussed in terms of how learning, plasticity, and sound detection are reflected in the auditory ERFs. The notion of the grand-averaged ERF is critically evaluated.

In this artifact, we document our publicly shared data set of our functional magnetic resonance imaging (fMRI) study on programmers. We have conducted an fMRI study with 19 participants observing program comprehension of short code snippets at varying complexity levels. We dissected four classes of code complexity metrics and their relationship to neuronal, behavioral, and subjective correlates of program comprehension. Our data corroborate that complexity metrics can—to a limited degree—explain programmers' cognition in program comprehension. In the paper on the fMRI study, we outline several follow-up experiments investigating fine-grained effects of code complexity and describe possible refinements to code complexity metrics. We view our conducted experiment as a starting point to link code complexity metrics to neural and behavioral correlates. To enable future research to continue this line of work, we aim to provide as much support as possible to conduct similar studies with this artifact.

Background: Researchers and practitioners have been using code complexity metrics for decades to predict how developers comprehend a program. While it is plausible and tempting to use code metrics for this purpose, their validity is debated, since they rely on simple code properties and rarely consider particularities of human cognition. Aims: We investigate whether and how code complexity metrics reflect difficulty of program comprehension. Method: We have conducted a functional magnetic resonance imaging (fMRI) study with 19 participants observing program comprehension of short code snippets at varying complexity levels. We dissected four classes of code complexity metrics and their relationship to neuronal, behavioral, and subjective correlates of program comprehension, overall analyzing more than 41 metrics. Results: While our data corroborate that complexity metrics can-to a limited degree-explain programmers' cognition in program comprehension, fMRI allowed us to gain insights into why some code properties are difficult to process. In particular, a code's textual size drives programmers' attention, and vocabulary size burdens programmers' working memory. Conclusion: Our results provide neuro-scientific evidence supporting warnings of prior research questioning the validity of code complexity metrics and pin down factors relevant to program comprehension. Future Work: We outline several follow-up experiments investigating fine-grained effects of code complexity and describe possible refinements to code complexity metrics.

Human learning is one of the main topics in psychology and cognitive neuroscience. The analysis of experimental data, e.g. from category learning experiments, is a major challenge due to confounding factors related to perceptual processing, feedback value, response selection, as well as inter-individual differences in learning progress due to differing strategies or skills. We use machine learning to investigate (Q1) how participants of an auditory category-learning experiment evolve towards learning, (Q2) how participant performance saturates and (Q3) how early we can differentiate whether a participant has learned the categories or not. We found that a Gaussian Mixture Model describes well the evolution of participant performance and serves as basis for identifying influencing factors of task configuration (Q1). We found early saturation trends (Q2) and that CatBoost, an advanced classification algorithm, can separate between participants who learned the categories and those who did not, well before the end of the learning session, without much degradation of separation quality (Q3). Our results show that machine learning can model participant dynamics, identify influencing factors of task design and performance trends. This will help to improve computational models of auditory category learning and define suitable time points for interventions into learning, e.g. by tutorial systems.

Programming research has entered the Neuroage.

Transcranial direct current stimulation (tDCS) is one of the most prominent non-invasive electrical brain stimulation method to alter neuronal activity as well as behavioral processes in cognitive and perceptual domains. However, the exact mode of action of tDCS-related cortical alterations is still unclear as the results of tDCS studies often do not comply with the somatic doctrine assuming that anodal tDCS enhances while cathodal tDCS decreases neuronal excitability. Changes in the regional cortical neurotransmitter balance within the stimulated cortex, measured by excitatory and inhibitory neurotransmitter levels, have the potential to provide direct neurochemical underpinnings of tDCS effects. Here we assessed tDCS-induced modulations of the neurotransmitter concentrations in the human auditory cortex (AC) by using magnetic resonance spectroscopy (MRS) at ultra-high-field (7 T). We quantified inhibitory gamma-amino butyric (GABA) concentration and excitatory glutamate (Glu) and compared changes in the relative concentration of GABA to Glu before and after tDCS application. We found that both, anodal and cathodal tDCS significantly increased the relative concentration of GABA to Glu with individual temporal specificity. Our results offer novel insights for a potential neurochemical mechanism that underlies tDCS-induced alterations of AC processing.