Maximizing reward guides how humans learn in a dynamically changing environment. However, the neural mechanisms underlying how reward promotes visual perceptual learning remain largely unknown. We trained 2 groups of participants (the Reward and the Control group) on a contrast categorization task on oriented Gabor patches in 1 visual quadrant for consecutive 5 sessions. Unknown to the participants in the Reward group, one of the orientations, which was presented but remained task-irrelevant during training, was paired with a monetary reward with 80% probability. Participants' orientation discrimination threshold in the Reward group improved for the reward-paired orientation in the trained visual quadrant, indicating visual perceptual learning of the task-irrelevant visual feature in association with reward. Learning also transferred to the untrained visual quadrants. No performance improvement was found in the Control group for either orientation. Surprisingly, we found that the integrity of the locus coeruleus modulates the amount of transfer of visual perceptual learning of the task-irrelevant visual feature: the higher the integrity of the locus coeruleus, the more likely learning of the task-irrelevant information will transfer to the untrained location. Our study provides important implications for the roles of locus coeruleus in reward-induced learning of irrelevant visual information.

Previous research suggests that action video game players (AVGPs) often outperform non-action video game players (NAVGPs) in cognitive tasks. This study compared the precision of visual short-term memory (VSTM) for motion direction between AVGPs and age- and gender-matched NAVGPs. Participants memorized the direction of random dot kinematograms (RDKs) presented sequentially (one to four per trial) and reproduced the direction of a probed RDK after either a short (0.5 s) or long (3 s) delay. Initial training ensured that all participants reached a predefined performance level with a single stimulus, with AVGPs requiring fewer training blocks to meet this criterion. While no significant group differences emerged at short delays, AVGPs showed significantly higher raw precision than NAVGPs in long-delay trials involving a single stimulus. However, this group difference did not reach significance in the corresponding precision parameter estimated by the Standard Mixture Model. To investigate memory-encoding strategies, we applied the resource-rational model (RRM), which formalizes the trade-off between behavioral accuracy and neural cost. Model estimates showed that NAVGPs placed greater weight on neural cost relative to behavioral benefits during encoding, particularly in long-delay trials, leading to reduced precision. In contrast, AVGPs allocated memory resources more efficiently, maintaining higher precision over extended intervals. These findings suggest that AVGPs adopt more effective encoding strategies, dynamically adjusting resource allocation to task demands. This study highlights the utility of resource-rational modeling for understanding cognitive performance differences linked to action video game experience. Future research could further explore how these strategies translate across different cognitive domains.

Contour erasure describes the phenomenon that after brief flicker adaptation at the edge of an object, the object disappears and is replaced by the background - highlighting the importance of edges in perceiving a surface. The underlying mechanism remains unknown. The current study investigates the characteristics and functional properties of contour erasure, and its relationship with related phenomena such as perceptual filling-in, forward masking, and contrast adaptation. We used a homogeneous disk as a target, and circles that corresponded to the outline of the target disk as the adapter. Using a two-alternative forced choice (2AFC) paradigm, each trial began with a counterphase flickering adapter, followed by the target randomly presented in one of the two locations. Participants indicated the target location with a button press. The target detection threshold elevation relative to the no adaptation condition was used as an index of the adaptation effect. We manipulated two spatial properties (eccentricity and the adapter size) plus three temporal properties (adapter flickering rate, adaptation duration, and interstimulus interval [ISI]). Results indicated that the adaptation effect increased with eccentricity, flickering rate (plateauing at 6 hertz [Hz]) and adaptation duration, but decreased with longer ISI and for adapter sizes that were larger than the target. The target threshold first increased then decreased as the adapter size decreased from that of the target, indicating a size tuning that is slightly smaller than the target. Our results indicate that contour erasure shares some of the key features of other well-known perceptual phenomena like filling in and contrast adaptation.

We here provide evidence for a benefit of congruent olfactory sensations during a challenging visual search task. Using a four-channel olfactometer, we exposed our participants to one of three suprathreshold fruit odorants (lemon, apple, or strawberry) or neutral room air (no odorant) while they searched for an image of a cued-target fruit presented among fruit distractors. Congruent odorants (e.g., exposure to a lemon scent while searching for a lemon among other fruits) led to greater performance levels and faster responses compared to trials where the participants were exposed to an odorant that was incongruent to the searched-for target fruit or to neutral room air, respectively. Post-hoc correlations across our n = 22 participants suggest that low performers in the baseline (no-odorant) search task benefited most from the congruent odorant-visual object coupling, whereas the visual search of high performers was impaired most by incongruent couplings. A control task points to large individual differences in olfactory discrimination across our healthy participants and their discriminative ability correlates with the amplitude and sign of the congruency effect. Our findings point to a multisensory congruency effect of odorant processing while participants visually search for a nutritional item among other food choices, suggesting that your nose knows what you are looking for.

This study aimed to investigate the impact of eccentric-vision training on population receptive field (pRF) estimates to provide insights into brain plasticity processes driven by practice. Fifteen participants underwent functional magnetic resonance imaging (fMRI) measurements before and after behavioral training on a visual crowding task, where the relative orientation of the opening (gap position: up/down, left/right) in a Landolt C optotype had to be discriminated in the presence of flanking ring stimuli. Drifting checkerboard bar stimuli were used for pRF size estimation in multiple regions of interest (ROIs): dorsal-V1 (dV1), dorsal-V2 (dV2), ventral-V1 (vV1), and ventral-V2 (vV2), including the visual cortex region corresponding to the trained retinal location. pRF estimates in V1 and V2 were obtained along eccentricities from 0.5° to 9°. Statistical analyses revealed a significant decrease of the crowding anisotropy index (p = 0.009) after training, indicating improvement on crowding task performance following training. Notably, pRF sizes at and near the trained location decreased significantly (p = 0.005). Dorsal and ventral V2 exhibited significant pRF size reductions, especially at eccentricities where the training stimuli were presented (p < 0.001). In contrast, no significant changes in pRF estimates were found in either vV1 (p = 0.181) or dV1 (p = 0.055) voxels. These findings suggest that practice on a crowding task can lead to a reduction of pRF sizes in trained visual cortex, particularly in V2, highlighting the plasticity and adaptability of the adult visual system induced by prolonged training.

The pervasive use of information technologies (IT) has tremendously benefited our daily lives. However, unpredicted technical breakdowns and errors can lead to the experience of stress, which has been termed technostress. It remains poorly understood how people dynamically respond to unpredicted system runtime errors occurring while interacting with the IT systems on a behavioral and neuronal level. To elucidate the mechanisms underlying such processes, we conducted a functional magnetic resonance imaging (fMRI) study in which 15 young adults solved arithmetic problems of three difficulty levels (easy, medium and hard) while two types of system runtime errors (problem errors and feedback errors) occurred in an unexpected manner. The problem error condition consisted of apparently defective displays of the arithmetic problem and the feedback error condition involved erroneous feedback. We found that the problem errors positively influenced participants' problem-solving performance at the high difficulty level (i.e., hard tasks) at the initial stage of the session, while feedback errors disturbed their performance. These dynamic behavioral changes are mainly associated with brain activation changes in the posterior cingulate and the default mode network, including the posterior cingulate cortex, the mPFC, the retrosplenial cortex and the parahippocampal gyrus. Our study illustrates the regulatory role of the posterior cingulate in coping with unpredicted errors as well as with dynamic changes in the environment.

We present a protocol to conduct functional magnetic resonance spectroscopy (fMRS) in human participants before, during, and after training on a visual task. We describe steps for participant setup, volume-of-interest placement, fMRS measurement, and post-scan tests. We discuss the design, analysis, and interpretation of fMRS experiments. This protocol can be adapted to investigate the dynamics of chief excitatory and inhibitory neurotransmitters (glutamate and g-aminobutyric acid, GABA, respectively) while participants perform or learn perceptual, motor, or cognitive tasks. For complete details on the use and execution of this protocol, please refer to Frank et al. (2022).1

Neuromodulatory signals such as reward and arousal modulate visual perceptual learning (VPL). Two hypotheses could explain how reward and arousal modulate VPL. A goal-dominance model predicts that reward or arousal enhance visual features relevant to the current goal and inhibit irrelevant features. Conversely, a state-dominance model predicts that reward or arousal enhance visual features irrespective of their relevance to the current goal. To test which of these models is consistent with the effects of reward and arousal on VPL, we trained three participant groups (Reward, Arousal and Control groups) over the course of 5 daily sessions on a VPL task, consisting of a Gabor patch presented in the upper left or lower right visual quadrant. The Gabor had one of two orientations (2.5 degree left or right tilted from vertical) and one of several contrast levels. Participants were instructed to categorize the Gabor based on contrast level, while maintaining central fixation (eye-tracking was conducted). The orientation of the Gabor was task-irrelevant. Unknown to the participants, monetary reward was paired 80% of the times with one of the orientations in the Reward group while an arousing beep was paired in the Arousal group. The reward group was instructed that reward was given when they maintained good fixation. No neuromodulatory signal was provided in the Control group. Before the first and after the final training session, participants performed an orientation discrimination task on different contrast levels in both quadrants. Participants’ performance decreased for the paired orientation in the Reward group, whereas performance enhanced for both the paired and unpaired orientations in the Arousal group. There were no performance changes in the Control group. These results indicate that the effects of reward on VPL are consistent with a goal-dominance model, whereas the effects of arousal on VPL are consistent with a state-dominance model.

The perception of coherent form configurations in natural scenes relies on the activity of early visual areas that respond to local orientation cues. Subsequently, high-level visual areas pool these local signals to construct a global representation of the initial visual input. However, it is still debated whether neurons in the early visual cortex respond also to global form features. Glass patterns (GPs) are visual stimuli employed to investigate local and global form processing and consist of randomly distributed dots pairs called dipoles arranged to form specific global configurations. In the current study, we used GPs and functional magnetic resonance imaging (fMRI) adaptation to reveal the visual areas that subserve the processing of oriented GPs. Specifically, we adapted participants to vertically oriented GP, then we presented test GPs having either the same or different orientations with respect to the adapting GP. We hypothesized that if local form features are processed exclusively by early visual areas and global form by higher-order visual areas, then the effect of visual adaptation should be more pronounced in higher tier visual areas as it requires global processing of the pattern. Contrary to this expectation, our results revealed that adaptation to GPs is robust in early visual areas (V1, V2, and V3), but not in higher tier visual areas (V3AB and V4v), suggesting that form cues in oriented GPs are primarily derived from local-processing mechanisms that originate in V1. Finally, adaptation to vertically oriented GPs causes a modification in the BOLD response within early visual areas, regardless of the relative orientations of the adapting and test stimuli, indicating a lack of orientation selectivity.

The interpretation of fMRI data in glioblastoma (GB) is challenging as these tumors exhibit specific hemodynamic processes which, together with malignancy, tumor volume and proximity to eloquent cortex areas, may lead to misinterpretations of fMRI signals. The aim of this study was to investigate if different radiologically defined GB tumor growth patterns may also influence the fMRI signal, activation pattern and functional connectivity differently. Sixty-four patients with left-hemispheric glioblastoma were included and stratified according to their radiologically defined tumor growth pattern into groups with a uniform (U-TGP) or diffuse tumor growth pattern (D-TGP). Task-based fMRI data were analyzed using SPM12 with the marsbar, LI and CONN toolboxes. The percent signal change and the laterality index were analyzed, as well as functional connectivity between 23 selected ROIs. Comparisons of both patient groups showed only minor non-significant differences, indicating that the tumor growth pattern is not a relevant influencing factor for fMRI signal. In addition to these results, signal reductions were found in areas that were not affected by the tumor underlining that a GB is not a localized but rather a systemic disease affecting the entire brain.

In this study, an automated 2D machine learning approach for fast and precise segmentation of MS lesions from multi-modal magnetic resonance images (mmMRI) is presented. The method is based on an U-Net like convolutional neural network (CNN) for automated 2D slice-based-segmentation of brain MRI volumes. The individual modalities are encoded in separate downsampling branches without weight sharing, to leverage the specific features. Skip connections input feature maps to multi-scale feature fusion (MSFF) blocks at every decoder stage of the network. Those are followed by multi-scale feature upsampling (MSFU) blocks which use the information about lesion shape and location. The CNN is evaluated on two publicly available datasets: The ISBI 2015 longitudinal MS lesion segmentation challenge dataset containing 19 subjects and the MICCAI 2016 MSSEG challenge dataset containing 15 subjects from various scanners. The proposed multi-input 2D architecture is among the top performing approaches in the ISBI challenge, to which open-access papers are available, is able to outperform state-of-the-art 3D approaches without additional post-processing, can be adapted to other scanners quickly, is robust against scanner variability and can be deployed for inference even on a standard laptop without a dedicated GPU.

Brief adaptation to a flickering outline of an object’s contour leads to its disappearance, which is referred to as contour erasure. Previous studies showed that contour erasure increased the contrast thresholds to targets with a homogeneous luminance contrast. Here, we used radial and concentric sinusoidal gratings as stimuli to investigate how contour erasure affects the detection of luminance modulation patterns, and how it interacts with lateral modulation. The target was a concentric grating with a crescent-shaped window (2-degree width and 2- to 8-degree length). The background patterns were concentric or radial gratings containing two blank regions (i.e., simulated scotomata) on the left and right of central fixation. The adapters were high-contrast 0.1-deg outlines of these blank regions. All patterns had a low (0.5 c/deg) spatial frequency. The target, scotomata, and the contour adapters had the same crescent shape and were presented at 5-deg eccentricity, left and right of fixation. In a two-alternative forced-choice (2-AFC) paradigm, two adapters first flickered at 3Hz for 1.5s, followed by the target appearing randomly in one of the two locations together with the background pattern for 83.3ms. The observers were asked to indicate the target location. We used a Bayesian adaptive staircase algorithm to estimate the target contrast threshold. Target thresholds decreased with target size and increased after contour adaptation. When the background was present, regardless of its orientation, target threshold increased with background contrast. The threshold was higher with the concentric background, suggesting an orientation-specific effect. The adaptation effect was absent with the concentric background whereas it was more prominent for the radial pattern with the low contrast adapter having a stronger effect than a high contrast one. Our results demonstrate that contour adaptation can also be induced for grating patterns with evidence for an interaction between contour-adaptation and lateral-modulation effects.

Visual attention can be employed to track independently moving objects. Such visual attentional tracking (VAT) is associated with fronto-parietal activation and also with cross-modal suppression of activation in the vestibular sensory system to minimize conflicts between vestibular and visual motion cues during VAT. Whether training to improve VAT modulates fronto-parietal activation and cross-modal vestibular suppression remains unclear. Such training might either lead to greater engagement of the visual attention system and thus greater fronto-parietal activation and cross-modal suppression, or it might be associated with a disengagement and thus decreased fronto-parietal activation and cross-modal suppression. We employed a VAT paradigm in which participants were asked to track four target disks among eight distractor disks for a total of 14s. Targets were only highlighted at trial start but were physically undistinguishable from distractors during tracking. At trial end participants were asked to indicate whether a randomly chosen disk was a target or a distractor. Participants (n=16) practiced the task over the course of 30min-long daily training sessions and reached a response accuracy of 90% correct, on average, after seven training sessions. This performance improvement was stable over the course of a one-month-long interval without any training. In a follow-up imaging experiment, a new group of participants (n=12) practiced the same VAT task behaviorally and was scanned with functional MRI and magnetic resonance spectroscopy before and after training. After training, fronto-parietal BOLD-activation decreased during VAT, while at the same time the magnitude of cross-modal suppression also decreased, as shown by a reduction of the concentration of GABA, a chief inhibitory neurotransmitter, during VAT in the vestibular cortex. Taken together, we find that training leads to long-lasting improvements of VAT that are associated with decreased fronto-parietal activation and cross-modal vestibular suppression, indicating that the visual attention system becomes less engaged with training of VAT.

In this study, an automated machine learning approach for the segmentation of MS lesions from multi- modal magnetic resonance images (mmMRI) is presented. The method is based on a U-Net like convolutional neural network (CNN) for 2D slice-based segmentation of 3D brain MRI volumes. The different modalities are encoded in sep- arate downsampling channels. The skip connections input feature maps to multi-scale feature fusion blocks at every stage of the network. These are followed by multi-scale feature upsampling blocks, which use the information from lesion shape and location. The CNN is evaluated on two publicly available datasets: The ISBI 2015 longitudinal MS lesion segmentation challenge dataset which contains 14 MS patients and the MICCAI 2016 MSSEG challenge dataset consisting of 15 MS patients. Regarding the ISBI Challenge, the proposed method was among the top performing ap- proaches to which open-access papers are available. The MICCAI dataset served to evaluate the robustness of the architecture against scanner variability and to show the improvements in performance after transfer learning.

Differences in the ability of students to judge images can be assessed by analyzing the individual preference order (ranking) of images. To gain insights into potential heterogeneity in judgement of visual abstraction among students, we combine Bradley-Terry preference modeling and model-based recursive partitioning. In an experiment a sample of 1,020 high-school students ranked five sets of images, three of which with respect to their level of visual abstraction. Additionally, 24 art experts and 25 novices were given the same task, while their eye movements were recorded. Results show that time spent on the task, the students' age, and self-reported interest in visual puzzles had significant influence on rankings. Fixation time of experts and novices revealed that both groups paid more attention to ambiguous images. The presented approach makes the underlying latent scale of visual judgments quantifiable.