Tyler Davis

What do differences between multi-voxel and univariate analysis mean? How subject-, voxel-, and trial-level variance impact fMRI analysis

Multi-voxel pattern analysis (MVPA) has led to major changes in how fMRI data are analyzed and interpreted. Many studies now report both MVPA results and results from standard univariate voxel-wise analysis, often with the goal of drawing different conclusions from each. Because MVPA results can be sensitive to latent multidimensional representations and processes whereas univariate voxel-wise analysis cannot, one conclusion that is often drawn when MVPA and univariate results differ is that the activation patterns underlying MVPA results contain a multidimensional code. In the current study, we conducted simulations to formally test this assumption. Our findings reveal that MVPA tests are sensitive to the magnitude of voxel-level variability in the effect of a condition within subjects, even when the same linear relationship is coded in all voxels. We also find that MVPA is insensitive to subject-level variability in mean activation across an ROI, which is the primary variance component of interest in many standard univariate tests. Together, these results illustrate that differences between MVPA and univariate tests do not afford conclusions about the nature or dimensionality of the neural code. Instead, targeted tests of the informational content and/or dimensionality of activation patterns are critical for drawing strong conclusions about the representational codes that are indicated by significant MVPA results.

Measuring neural representations with fMRI: practices and pitfalls

Recently, there has been a dramatic increase in the number of functional magnetic resonance imaging studies seeking to answer questions about how the brain represents information. Representational questions are of particular importance in connecting neuroscientific and cognitive levels of analysis because it is at the representational level that many formal models of cognition make distinct predictions. This review discusses techniques for univariate, adaptation, and multivoxel analysis, and how they have been used to answer questions about content specificity in different regions of the brain, how this content is organized, and how representations are shaped by and contribute to cognitive processes. Each of the analysis techniques makes different assumptions about the underlying neural code and thus differ in how they can be applied to specific questions. We also discuss the many pitfalls of representational analysis, from the flexibility in data analysis pipelines to emergent nonrepresentational relationships that can arise between stimuli in a task.

The impact of study design on pattern estimation for single-trial multivariate pattern analysis.

A prerequisite for a pattern analysis using functional magnetic resonance imaging (fMRI) data is estimating the patterns from time series data, which then are input into the pattern analysis. Here we focus on how the combination of study design (order and spacing of trials) with pattern estimator impacts the Type I error rate of the subsequent pattern analysis. When Type I errors are inflated, the results are no longer valid, so this work serves as a guide for designing and analyzing MVPA studies with controlled false positive rates. The MVPA strategies examined are pattern classification and similarity, utilizing single trial activation patterns from the same functional run. Primarily focusing on the Least Squares Single and Least Square All pattern estimators, we show that collinearities in the models, along with temporal autocorrelation, can cause false positive correlations between activation pattern estimates that adversely impact the false positive rates of pattern similarity and classification analyses. It may seem intuitive that increasing the interstimulus interval (ISI) would alleviate this issue, but remaining weak correlations between activation patterns persist and have a strong influence in pattern similarity analyses. Pattern similarity analyses using only activation patterns estimated from the same functional run of data are susceptible to inflated false positives unless trials are randomly ordered, with a different randomization for each subject. In other cases, where there is any structure to trial order, valid pattern similarity analysis results can only be obtained if similarity computations are restricted to pairs of activation patterns from independent runs. Likewise, for pattern classification, false positives are minimized when the testing and training sets in cross validation do not contain patterns estimated from the same run.

Striatal and Hippocampal Entropy and Recognition Signals in Category Learning: Simultaneous Processes Revealed by Model-Based fMRI

Category learning is a complex phenomenon that engages multiple cognitive processes, many of which occur simultaneously and unfold dynamically over time. For example, as people encounter objects in the world, they simultaneously engage processes to determine their fit with current knowledge structures, gather new information about the objects, and adjust their representations to support behavior in future encounters. Many techniques that are available to understand the neural basis of category learning assume that the multiple processes that subserve it can be neatly separated between different trials of an experiment. Model-based functional magnetic resonance imaging offers a promising tool to separate multiple, simultaneously occurring processes and bring the analysis of neuroimaging data more in line with category learnings dynamic and multifaceted nature. We use model-based imaging to explore the neural basis of recognition and entropy signals in the medial temporal lobe and striatum that are engaged while participants learn to categorize novel stimuli. Consistent with theories suggesting a role for the anterior hippocampus and ventral striatum in motivated learning in response to uncertainty, we find that activation in both regions correlates with a model-based measure of entropy. Simultaneously, separate subregions of the hippocampus and striatum exhibit activation correlated with a model-based recognition strength measure. Our results suggest that model-based analyses are exceptionally useful for extracting information about cognitive processes from neuroimaging data. Models provide a basis for identifying the multiple neural processes that contribute to behavior, and neuroimaging data can provide a powerful test bed for constraining and testing model predictions.

Memory for Category Information Is Idealized Through Contrast With Competing Options

We suggest that human category formation relies on contrastive learning mechanisms that seek to reduce prediction error. In keeping with this view, manipulating category contrast leads to systematic distortions in people’s memory for category information. Simply by changing the basis of comparison (i.e., the available response options), we can systematically distort people’s perceptions of novel, energy-source, and political categories. Our proposal explains perceived variations in category members’ typicality, including cases in which average items are judged as highly typical and cases in which extreme or ideal members are judged as highly typical of the category. Although straightforward, our account spans findings from studies in goal-derived, cross-cultural, and object-based categorization and suggests ways in which society’s perception of key issues is distorted by political discourse.

Global neural pattern similarity as a common basis for categorization and recognition memory.

Familiarity, or memory strength, is a central construct in models of cognition. In previous categorization and long-term memory research, correlations have been found between psychological measures of memory strength and activation in the medial temporal lobes (MTLs), which suggests a common neural locus for memory strength. However, activation alone is insufficient for determining whether the same mechanisms underlie neural function across domains. Guided by mathematical models of categorization and long-term memory, we develop a theory and a method to test whether memory strength arises from the global similarity among neural representations. In human subjects, we find significant correlations between global similarity among activation patterns in the MTLs and both subsequent memory confidence in a recognition memory task and model-based measures of memory strength in a category learning task. Our work bridges formal cognitive theories and neuroscientific models by illustrating that the same global similarity computations underlie processing in multiple cognitive domains. Moreover, by establishing a link between neural similarity and psychological memory strength, our findings suggest that there may be an isomorphism between psychological and neural representational spaces that can be exploited to test cognitive theories at both the neural and behavioral levels.

Anticipatory emotions in decision tasks: Covert markers of value or attentional processes?

Anticipatory emotions precede behavioral outcomes and provide a means to infer interactions between emotional and cognitive processes. A number of theories hold that anticipatory emotions serve as inputs to the decision process and code the value or risk associated with a stimulus. We argue that current data do not unequivocally support this theory. We present an alternative theory whereby anticipatory emotions reflect the outcome of a decision process and serve to ready the subject for new information when making an uncertain response. We test these two accounts, which we refer to as emotions-as-input and emotions-as-outcome, in a task that allows risky stimuli to be dissociated from uncertain responses. We find that emotions are associated with responses as opposed to stimuli. This finding is contrary to the emotions-as-input perspective as it shows that emotions arise from decision processes.

Two pathways to stimulus encoding in category learning

Category learning theorists tacitly assume that stimuli are encoded by a single pathway. Motivated by theories of object recognition, we evaluated a dual-pathway account of stimulus encoding. The part-based pathway establishes mappings between sensory input and symbols that encode discrete stimulus features, whereas the image-based pathway applies holistic templates to sensory input. Our experiments used rule-plus-exception structures, in which one exception item in each category violates a salient regularity and must be distinguished from other items. In Experiment 1, we found discrete representations to be crucial for recognition of exceptions following brief training. Experiments 2 and 3 involved multisession training regimens designed to encourage either part- or image-based encoding. We found that both pathways are able to support exception encoding, but have unique characteristics. We speculate that one advantage of the part-based pathway is the ability to generalize across domains, whereas the image-based pathway provides faster and more effortless recognition.

From Concrete Examples to Abstract Relations: The Rostrolateral Prefrontal Cortex Integrates Novel Examples into Relational Categories

Abstract The ability to form relational categories for objects that share few features in common is a hallmark of human cognition. For example, anything that can play a preventative role, from a boulder to poverty, can be a “barrier.” However, neurobiological research has focused solely on how people acquire categories defined by features. The present functional magnetic resonance imaging study examines how relational and feature‐based category learning compare in well‐matched learning tasks. Using a computational model‐based approach, we observed a cluster in left rostrolateral prefrontal cortex (rlPFC) that tracked quantitative predictions for the representational distance between test and training examples during relational categorization. Contrastingly, medial and dorsal PFC exhibited graded activation that tracked decision evidence during both feature‐based and relational categorization. The results suggest that rlPFC computes an alignment signal that is critical for integrating novel examples during relational categorization whereas other PFC regions support more general decision functions.

The Evaluative Role of Rostrolateral Prefrontal Cortex in Rule-Based Category Learning

&NA; Category learning is a critical neurobiological function that allows organisms to simplify a complex world. Rostrolateral prefrontal cortex (rlPFC) is often active in neurobiological studies of category learning; however, the specific role this region serves in category learning remains uncertain. Previous category learning studies have hypothesized that the rlPFC is involved in switching between rules, whereas others have emphasized rule abstraction and evaluation. We aimed to clarify the role of rlPFC in category learning and dissociate switching and evaluation accounts using two common types of category learning tasks: matching and classification. The matching task involved matching a reference stimulus to one of four target stimuli. In the classification task, participants were shown a single stimulus and learned to classify it into one of two categories. Matching and classification are similar but place different demands on switching and evaluation. In matching, a rule can be known with certainty after a single correct answer. In classification, participants may need to evaluate evidence for a rule even after an initial correct response. This critical difference allows isolation of evaluative functions from switching functions. If the rlPFC is primarily involved in switching between representations, it should cease to be active once participants settle on a given rule in both tasks. If the rlPFC is involved in rule evaluation, its activation should persist in the classification task, but not matching. The results revealed that rlPFC activation persisted into correct trials in classification, but not matching, suggesting that it continues to be involved in the evaluations of evidence for a rule even after participants have arrived at the correct rule. HighlightsDifferences between rule‐based matching and classification tasks were highlighted.Rostrolateral prefrontal cortex remains engaged past when participants actively switch between rules.Rostrolateral prefrontal cortex is involved in evaluation of evidence for a rule in rule‐based category learning tasks.