Humans combine value learning and hypothesis testing strategically in multi-dimensional probabilistic reward learning.

Realistic and complex decision tasks often allow for many possible solutions. How do we find the correct one? Introspection suggests a process of trying out solutions one after the other until success. However, such methodical serial testing may be too slow, especially in environments with noisy feedback. Alternatively, the underlying learning process may involve implicit reinforcement learning that learns about many possibilities in parallel. Here we designed a multi-dimensional probabilistic active-learning task tailored to study how people learn to solve such complex problems. Participants configured three-dimensional stimuli by selecting features for each dimension and received probabilistic reward feedback. We manipulated task complexity by changing how many feature dimensions were relevant to maximizing reward, as well as whether this information was provided to the participants. To investigate how participants learn the task, we examined models of serial hypothesis testing, feature-based reinforcement learning, and combinations of the two strategies. Model comparison revealed evidence for hypothesis testing that relies on reinforcement-learning when selecting what hypothesis to test. The extent to which participants engaged in hypothesis testing depended on the instructed task complexity: people tended to serially test hypotheses when instructed that there were fewer relevant dimensions, and relied more on gradual and parallel learning of feature values when the task was more complex. This demonstrates a strategic use of task information to balance the costs and benefits of the two methods of learning.

BrainIAK: The Brain Imaging Analysis Kit.

Functional magnetic resonance imaging (fMRI) offers a rich source of data for studying the neural basis of cognition. Here, we describe the Brain Imaging Analysis Kit (BrainIAK), an open-source, free Python package that provides computationally optimized solutions to key problems in advanced fMRI analysis. A variety of techniques are presently included in BrainIAK: intersubject correlation (ISC) and intersubject functional connectivity (ISFC), functional alignment via the shared response model (SRM), full correlation matrix analysis (FCMA), a Bayesian version of representational similarity analysis (BRSA), event segmentation using hidden Markov models, topographic factor analysis (TFA), inverted encoding models (IEMs), an fMRI data simulator that uses noise characteristics from real data (fmrisim), and some emerging methods. These techniques have been optimized to leverage the efficiencies of high-performance compute (HPC) clusters, and the same code can be se amlessly transferred from a laptop to a cluster. For each of the aforementioned techniques, we describe the data analysis problem that the technique is meant to solve and how it solves that problem; we also include an example Jupyter notebook for each technique and an annotated bibliography of papers that have used and/or described that technique. In addition to the sections describing various analysis techniques in BrainIAK, we have included sections describing the future applications of BrainIAK to real-time fMRI, tutorials that we have developed and shared online to facilitate learning the techniques in BrainIAK, computational innovations in BrainIAK, and how to contribute to BrainIAK. We hope that this manuscript helps readers to understand how BrainIAK might be useful in their research.

Incorporating structured assumptions with probabilistic graphical models in fMRI data analysis.

With the wide adoption of functional magnetic resonance imaging (fMRI) by cognitive neuroscience researchers, large volumes of brain imaging data have been accumulated in recent years. Aggregating these data to derive scientific insights often faces the challenge that fMRI data are high-dimensional, heterogeneous across people, and noisy. These challenges demand the development of computational tools that are tailored both for the neuroscience questions and for the properties of the data. We review a few recently developed algorithms in various domains of fMRI research: fMRI in naturalistic tasks, analyzing full-brain functional connectivity, pattern classification, inferring representational similarity and modeling structured residuals. These algorithms all tackle the challenges in fMRI similarly: they start by making clear statements of assumptions about neural data and existing domain knowledge, incorporate those assumptions and domain knowledge into probabilistic graphical models, and use those models to estimate properties of interest or latent structures in the data. Such approaches can avoid erroneous findings, reduce the impact of noise, better utilize known properties of the data, and better aggregate data across groups of subjects. With these successful cases, we advocate wider adoption of explicit model construction in cognitive neuroscience. Although we focus on fMRI, the principle illustrated here is generally applicable to brain data of other modalities.

Facilitating open-science with realistic fMRI simulation: validation and application.

With advances in methods for collecting and analyzing fMRI data, there is a concurrent need to understand how to reliably evaluate and optimally use these methods. Simulations of fMRI data can aid in both the evaluation of complex designs and the analysis of data. We present fmrisim, a new Python package for standardized, realistic simulation of fMRI data. This package is part of BrainIAK: a recently released open-source Python toolbox for advanced neuroimaging analyses. We describe how to use fmrisim to extract noise properties from real fMRI data and then create a synthetic dataset with matched noise properties and a user-specified signal. We validate the noise generated by fmrisim to show that it can approximate the noise properties of real data. We further show how fmrisim can help researchers find the optimal design in terms of power. The fmrisim package holds promise for improving the design of fMRI experiments, which may facilitate both the pre-registration of such experiments as well as the analysis of fMRI data.

Representational structure or task structure? Bias in neural representational similarity analysis and a Bayesian method for reducing bias.

The activity of neural populations in the brains of humans and animals can exhibit vastly different spatial patterns when faced with different tasks or environmental stimuli. The degrees of similarity between these neural activity patterns in response to different events are used to characterize the representational structure of cognitive states in a neural population. The dominant methods of investigating this similarity structure first estimate neural activity patterns from noisy neural imaging data using linear regression, and then examine the similarity between the estimated patterns. Here, we show that this approach introduces spurious bias structure in the resulting similarity matrix, in particular when applied to fMRI data. This problem is especially severe when the signal-to-noise ratio is low and in cases where experimental conditions cannot be fully randomized in a task. We propose Bayesian Representational Similarity Analysis (BRSA), an alternative method for computing representational similarity, in which we treat the covariance structure of neural activity patterns as a hyper-parameter in a generative model of the neural data. By marginalizing over the unknown activity patterns, we can directly estimate this covariance structure from imaging data. This method offers significant reductions in bias and allows estimation of neural representational similarity with previously unattained levels of precision at low signal-to-noise ratio, without losing the possibility of deriving an interpretable distance measure from the estimated similarity. The method is closely related to Pattern Component Model (PCM), but instead of modeling the estimated neural patterns as in PCM, BRSA models the imaging data directly and is suited for analyzing data in which the order of task conditions is not fully counterbalanced. The probabilistic framework allows for jointly analyzing data from a group of participants. The method can also simultaneously estimate a signal-to-noise ratio map that shows where the learned representational structure is supported more strongly. Both this map and the learned covariance matrix can be used as a structured prior for maximum a posteriori estimation of neural activity patterns, which can be further used for fMRI decoding. Our method therefore paves the way towards a more unified and principled analysis of neural representations underlying fMRI signals. We make our tool freely available in Brain Imaging Analysis Kit (BrainIAK).

Real time functional MRI training to decrease motion in imaging studies: Lack of significant improvement.

Functional magnetic resonance imaging (fMRI) is widely used to study brain circuitry in healthy controls and in psychiatry. A major problem of fMRI studies is motion, which affects the quality of images, is a major source of noise, and can confound data if, for example, the experimental groups move differently. Despite continual reminders to experimental subjects about keeping still, however, movement in the scanner remains a problem. The authors hypothesized that showing head movement during a scanning session may help subjects learn how to keep their head still. The authors scanned subjects and displayed in real time a plot of head movement that had three regions. The authors found, in a limited sample, that the improvements were marginal and inconsistent. Thus, they concluded that this strategy, even if likely to work for some people, is probably not sufficiently successful to be implemented at this time.

Human Orbitofrontal Cortex Represents a Cognitive Map of State Space.

Although the orbitofrontal cortex (OFC) has been studied intensely for decades, its precise functions have remained elusive. We recently hypothesized that the OFC contains a "cognitive map" of task space in which the current state of the task is represented, and this representation is especially critical for behavior when states are unobservable from sensory input. To test this idea, we apply pattern-classification techniques to neuroimaging data from humans performing a decision-making task with 16 states. We show that unobservable task states can be decoded from activity in OFC, and decoding accuracy is related to task performance and the occurrence of individual behavioral errors. Moreover, similarity between the neural representations of consecutive states correlates with behavioral accuracy in corresponding state transitions. These results support the idea that OFC represents a cognitive map of task space and establish the feasibility of decoding state representations in humans using non-invasive neuroimaging.

Duration estimates within a modality are integrated sub-optimally.

Perceived duration can be influenced by various properties of sensory stimuli. For example, visual stimuli of higher temporal frequency are perceived to last longer than those of lower temporal frequency. How does the brain form a representation of duration when each of two simultaneously presented stimuli influences perceived duration in different way? To answer this question, we investigated the perceived duration of a pair of dynamic visual stimuli of different temporal frequencies in comparison to that of a single visual stimulus of either low or high temporal frequency. We found that the duration representation of simultaneously occurring visual stimuli is best described by weighting the estimates of duration based on each individual stimulus. However, the weighting performance deviates from the prediction of statistically optimal integration. In addition, we provided a Bayesian account to explain a difference in the apparent sensitivity of the psychometric curves introduced by the order in which the two stimuli are displayed in a two-alternative forced-choice task.

Perceived duration is reduced by repetition but not by high-level expectation.

A repeated stimulus is judged as briefer than a novel one. It has been suggested that this duration illusion is an example of a more general phenomenon-namely that a more expected stimulus is judged as briefer than a less expected one. To test this hypothesis, we manipulated high-level expectation through the probability of a stimulus sequence, through the regularity of the preceding stimuli in a sequence, or through whether a stimulus violates an overlearned sequence. We found that perceived duration is not reduced by these types of expectation. Repetition of stimuli, on the other hand, consistently reduces perceived duration across our experiments. In addition, the effect of stimulus repetition is constrained to the location of the repeated stimulus. Our findings suggest that estimates of subsecond duration are largely the result of low-level sensory processing.