Biology subjects

Turk-Browne, N. B.

Publications and source records attributed to Turk-Browne, N. B..

51 records · Page 3Linked to original sources

bioRxiv2020

Emergence and organization of adult brain function throughout child development

Adult cognitive neuroscience has guided the study of human brain development by identifying regions associated with cognitive functions at maturity. The activity, connectivity, and structure of a region can be compared across ages to characterize the developmental trajectory of the corresponding function. However, observed developmental differences may not only reflect the maturation of the function but also its organization across the brain. That is, a function may be mature in children but supported by different brain regions and thus underestimated by focusing on adult regions. To test these possibilities, we investigated the presence, maturity, and localization of adult functions in children using probabilistic shared response modeling, a machine learning approach for functional alignment. After learning a lower-dimensional feature space from fMRI activity as adults watched a movie, we translated these shared features into the anatomical brain space of children 3-12 years old. To evaluate functional maturity, we correlated this reconstructed activity with the childrens actual fMRI activity as they watched the same movie. We found reliable correlations throughout cortex, even in the youngest children. The strength of the correlation in the precuneus, inferior frontal gyrus, and lateral occipital cortex increased over development and predicted chronological age. These age-related changes were driven by three types of developmental trajectories across distinct features of adult function: emergence from absence to presence, consistency in anatomical expression, and reorganization from one anatomical region to another. This data-driven approach to studying brain-wide function during naturalistic perception provides an abstract description of cognitive development throughout childhood. Significance StatementWhen watching a movie, your brain processes many types of information--plotlines, characters, locations, etc. A child watching this movie receives the same input, but some of their cognitive abilities (e.g., motion detection) are more developed than others (e.g., emotional reasoning). Beyond anatomical differences, when does the child brain begin to function like an adult brain? We used a data-driven approach to extract different aspects of brain activity from adults while they watched a movie during fMRI. We then predicted what the brain activity of a child would look like if they had processed the movie the same way. Comparing this prediction with actual brain activity from children allowed us to track the development of human brain function.

neuroscience↗

bioRxiv2020

Searching through functional space reveals distributed visual, auditory, and semantic coding in the human brain

The extent to which brain functions are localized or distributed is a foundational question in neuroscience. In the human brain, common fMRI methods such as cluster correction, atlas parcellation, and anatomical searchlight are biased by design toward finding localized representations. Here we introduce the functional searchlight approach as an alternative to anatomical searchlight analysis, the most commonly used exploratory multivariate fMRI technique. Functional searchlight removes any anatomical bias by grouping voxels based only on functional similarity and ignoring anatomical proximity. We report evidence that visual and auditory features from deep neural networks and semantic features from a natural language processing model are more widely distributed across the brain than previously acknowledged. This approach provides a new way to evaluate and constrain computational models with brain activity and pushes our understanding of human brain function further along the spectrum from strict modularity toward distributed representation.

neuroscience↗

bioRxiv2020

How to read a baby's mind: Re-imagining fMRI for awake, behaving infants

Thousands of functional magnetic resonance imaging (fMRI) studies have provided important insight into the human brain. However, only a handful of these studies tested infants while they were awake, because of the significant and unique methodological challenges involved. We report our efforts over the past five years to address these challenges, with the goal of creating methods for infant fMRI that can reveal the inner workings of the developing, preverbal mind. We use these methods to collect and analyze two fMRI datasets obtained from infants during cognitive tasks, released publicly with this paper. In these datasets, we explore data quantity and quality, task-evoked activity, and preprocessing decisions to derive and evaluate recommendations for infant fMRI. We disseminate these methods by sharing two software packages that integrate infant-friendly cognitive tasks and behavioral monitoring with fMRI acquisition and analysis. These resources make fMRI a feasible and accessible technique for cognitive neuroscience in human infants.

neuroscience↗

bioRxiv2019

Learning not to remember: How predicting the future impairs encoding of the present

Memory is typically thought of as enabling reminiscence about past experiences. However, memory also informs and guides processing of future experiences. These two functions of memory are often at odds: remembering specific experiences from the past requires storing idiosyncratic properties that define particular moments in space and time, but by definition such properties will not be shared with similar situations in the future and thus may not be applicable to future situations. We discovered that, when faced with this conflict, the brain prioritizes prediction over encoding. Behavioral tests of recognition and source recall showed that items allowing for prediction of what will appear next based on learned regularities were less likely to be encoded into memory. Brain imaging revealed that the hippocampus was responsible for this interference between statistical learning and episodic memory. The more that the hippocampus predicted the category of an upcoming item, the worse the current item was encoded. This competition may serve an adaptive purpose, focusing encoding on experiences for which we do not yet have a predictive model.

neuroscience↗

bioRxiv2019

Functions of ventral visual cortex after bilateral hippocampal loss

Repeated stimuli elicit attenuated responses in visual cortex relative to novel stimuli. This adaptation phenomenon can be considered a form of rapid learning and a signature of perceptual memory. Adaptation occurs not only when a stimulus is repeated immediately, but also when there is a lag in terms of time and other intervening stimuli before the repetition. But how does the visual system keep track of which stimuli are repeated, especially after long delays and many intervening stimuli? We hypothesized that the hippocampus supports long-lag adaptation, given that it learns from single experiences, maintains information over delays, and sends feedback to visual cortex. We tested this hypothesis with fMRI in an amnesic patient, LSJ, who has encephalitic damage to the medial temporal lobe resulting in complete bilateral hippocampal loss. We measured adaptation at varying time lags between repetitions in functionally localized visual areas that were intact in LSJ. We observed that these areas track information over a few minutes even when the hippocampus is unavailable. Indeed, LSJ and controls were identical when attention was directed away from the repeating stimuli: adaptation occurred for lags up to three minutes, but not six minutes. However, when attention was directed toward stimuli, controls now showed an adaptation effect at six minutes but LSJ did not. These findings suggest that visual cortex can support one-shot perceptual memories lasting for several minutes but that the hippocampus is necessary for adaptation in visual cortex after longer delays when stimuli are task-relevant.

neuroscience↗

bioRxiv2019

Content-based dissociation of hippocampal involvement in prediction

It has recently become clear that one of the key functions of the hippocampus is to predict future inputs. In line with this, previous research has revealed prediction-related signals in the hippocampus for complex visual objects, such as fractals and abstract shapes. Based on this, it has been suggested that the hippocampus may generate perceptual expectations, especially when relying on rapidly learned predictive associations between arbitrary stimuli. However, it is currently unknown whether the hippocampus implements general-purpose computations that subserve all associative predictions, regardless of stimulus properties, or whether the involvement of the hippocampus is stimulus-dependent. To investigate this, we exposed male and female human participants to complex auditory cues predicting either the shape of a complex object (Experiment 1) or the orientation of a simple line grating (Experiment 2). We measured brain activity using high-resolution functional magnetic resonance imaging (fMRI), in combination with inverted encoding models to reconstruct shape and orientation representations in visual cortex and the hippocampus. Our results revealed that expectations about shape and orientation evoked distinct representations in the hippocampus. For complex shapes, the hippocampus represented which shape was expected, potentially serving as a source of top-down predictions. In contrast, for simple gratings, the hippocampus represented only unexpected orientations, more reminiscent of a prediction error. We discuss several potential explanations for this dissociation, concluding that the computational role of the hippocampus in predictive processing depends upon the nature and complexity of stimuli.\n\nSignificance StatementTo deal with the noisy and ambiguous sensory signals received by our brain, it is crucial to use prior knowledge of the world to guide perception. Previous research suggests that the hippocampus is involved in predicting upcoming visual stimuli based on prior knowledge. In the current study, we show that hippocampal prediction is specific to expectations of complex objects, whereas for simple features the hippocampus generates an opposite prediction error signal instead. These findings demonstrate that the computational role of the hippocampus can be content-sensitive and refine our understanding of the involvement of memory systems in predictive processing.

neuroscience↗

bioRxiv2018

Hippocampal-neocortical interactions sharpen over time for predictive actions

When an action is familiar, we are able to anticipate how it will change the state of the world. These expectations can result from retrieval of action-outcome associations in the hippocampus and the reinstatement of anticipated outcomes in visual cortex. How does this role for the hippocampus in action-based prediction change over time? We used high-resolution fMRI and a dual-training behavioral paradigm to examine how the hippocampus interacts with visual cortex during predictive and nonpredictive actions learned either three days earlier or immediately before the scan. Just-learned associations led to comparable background connectivity between the hippocampus and V1/V2, regardless of whether actions predicted outcomes. However, three-day-old associations led to stronger background connectivity and greater differentiation between neural patterns for predictive vs. nonpredictive actions. Hippocampal prediction may initially reflect indiscriminate binding of co-occurring of events, with action information pruning weaker associations and leading to more selective and accurate predictions over time.

neuroscience↗

bioRxiv2018

Focusing on what matters: Modulation of the human hippocampus by relational attention

Hippocampal episodic memory is fundamentally relational, consisting of links between events and the spatial and temporal contexts in which they occurred. Such relations are also important over much shorter time periods, during online visual perception. For example, how do we assess the relative spatial positions of objects, their temporal order, or the relationship between their features? Here, we investigate the role of the hippocampus in such online relational processing by manipulating visual attention to different kinds of relations in a dynamic display. While undergoing high-resolution fMRI, participants viewed two images in rapid succession on each trial and performed one of three relational tasks, judging the images relative: spatial positions, temporal onsets, or sizes. As a control, they sometimes also judged whether one image was tilted, irrespective of the other; this served as a baseline item task with no demands on relational processing. All hippocampal regions of interest (CA1, CA2/3/DG, subiculum) showed reliable deactivation when participants attended to relational vs. item information. Attention to temporal relations was associated with more robust deactivation than the other conditions. One possible interpretation of such deactivation is that it reflects hippocampal disengagement. If true, there should be reduced information content and noisier, less reliable patterns of activity in the hippocampus for the temporal vs. other tasks. Instead, analyses of multivariate activity patterns revealed more stable hippocampal representations in the temporal task. Additional analyses showed that this increased pattern similarity was not simply a reflection of the lower univariate activity. Thus, the hippocampus differentiates between relational and item processing even during online visual perception, and its representations of temporal relations in particular are robust and stable. Together, these findings suggest that the relational computations of the hippocampus, known to be important for memory, extend beyond this purpose, enabling the rapid online extraction of relational information in visual perception.

neuroscience↗

bioRxiv2018

Neurofeedback helps to reveal a relationship between context reinstatement and memory retrieval

Theories of mental context and memory posit that successful mental context reinstatement enables better retrieval of memories from the same context, at the expense of memories from other contexts. To test this hypothesis, we had participants study lists of words, interleaved with task-irrelevant images from one category (e.g., scenes). Following encoding, participants were cued to mentally reinstate the context associated with a particular list, by thinking about the images that had appeared between the words. We measured context reinstatement with fMRI, and related this to performance on a free recall test that followed immediately afterwards. To increase sensitivity, we used a closed-loop neurofeedback procedure, whereby higher levels of context reinstatement (measured neurally) elicited increased visibility of the images from the studied context onscreen. Our goal was to create a positive feedback loop that amplified small fluctuations in mental context reinstatement, making it easier to experimentally detect a relationship between context reinstatement and recall. As predicted, we found that higher levels of neural context reinstatement were associated with better recall of words from the reinstated context, and worse recall of words from a different context. In a second experiment, we assessed the role of neurofeedback in identifying this brain-behavior relationship by presenting context images again but manipulating whether their visibility depended on neural context reinstatement. When neurofeedback was removed, the relationship between context reinstatement and memory retrieval disappeared. Together, these findings demonstrate a clear effect of context reinstatement on memory recall and suggest that neurofeedback can be a useful tool for characterizing brain-behavior relationships.\n\nAbbreviated titleNeurofeedback context

neuroscience↗

bioRxiv2018

Stimulus prediction in the hippocampus

Perception can be cast as a process of inference, in which bottom-up signals are combined with top-down predictions in sensory systems. However, the source of these top-down predictions, especially when complex and multisensory, remains largely unknown. We hypothesised that the hippocampus -- which rapidly learns arbitrary associations and has bidirectional connections with sensory systems -- may be involved. We exposed humans to auditory cues predicting visual shapes, while measuring high-resolution fMRI signals in visual cortex and the hippocampus. Using multivariate reconstruction methods, we discovered a dissociation between these regions: representations in visual cortex were dominated by whichever shape was presented, whereas representations in the hippocampus (CA3 and subiculum, but not CA1) reflected only which shape was predicted by the cue. The strength of hippocampal predictions correlated across participants with the amount of expectation-related facilitation in visual cortex. These findings are consistent with the possibility that the hippocampus supplies predictions to sensory systems.

neuroscience↗

bioRxiv2017

Learning naturalistic temporal structure in the posterior medial network

The posterior medial network is at the apex of a temporal integration hierarchy in the brain, integrating information over many seconds of viewing intact, but not scrambled, movies. This has been interpreted as an effect of temporal structure. Such structure in movies depends on pre-existing event schemas, but temporal structure can also arise de novo from learning. Here we examined the relative role of schema-consistent temporal structure and arbitrary but consistent temporal structure on the human posterior medial network. We tested whether, with repeated viewing, the network becomes engaged by scrambled movies with temporal structure. Replicating prior studies, posterior medial regions were immediately locked to stimulus structure upon exposure to intact but not scrambled movies. However, for temporally structured scrambled movies, functional coupling within the network increased across stimulus repetitions, rising to the level of intact movies. Thus, temporal structure is a key determinant of network dynamics and function.

neuroscience↗

bioRxiv2017

Perceptual decisions result from the continuous accumulation of memory and sensory evidence

Expectations can inform fast, accurate decisions. But what informs expectations? Here we test the hypothesis that expectations are set by dynamic inference from memory. Participants performed a cue-guided perceptual decision task with independently-varying memory and sensory evidence. Cues established expectations by reminding participants of past stimulus-stimulus pairings, which predicted the likely target in a subsequent noisy image stream. Participants responses used both memory and sensory information, in accordance to their relative reliability. Formal model comparison showed that the sensory inference was best explained when its parameters were set dynamically at each trial by evidence sampled from memory. Supporting this model, neural pattern analysis revealed that responses to the probe were modulated by the specific content and fidelity of memory reinstatement that occurred before the probe appeared. Together, these results suggest that perceptual decisions arise from the continuous sampling of memory and sensory evidence.

neuroscience↗

bioRxiv2017

Neural Overlap In Item Representations Across Episodes Impairs Context Memory

We frequently encounter the same item in different contexts, and when that happens, memories of earlier encounters can get reactivated in the brain. Here we examined how these existing memories are changed as a result of such reactivation. We hypothesized that when an items initial and subsequent neural representations overlap, this allows the initial item to become associated with novel contextual information, interfering with later retrieval of the initial context. That is, we predicted a negative relationship between representational similarity across repeated experiences of an item and subsequent source memory for the initial context. We tested this hypothesis in an fMRI study, in which objects were presented multiple times during different tasks. We measured the similarity of the neural patterns in lateral occipital cortex that were elicited by the first and second presentations of objects, and related this neural overlap score to source memory in a subsequent test. Consistent with our hypothesis, greater item-specific pattern similarity was linked to worse source memory for the initial task. Our findings suggest that the influence of novel experiences on an existing context memory depends on how reliably a shared component (i.e., same item) is represented across these episodes.

neuroscience↗

bioRxiv2017

Common object representations for visual production and recognition

Production and comprehension have long been viewed as inseparable components of language. The study of vision, by contrast, has centered almost exclusively on comprehension. Here we investigate drawing -- the most basic form of visual production. How do we convey concepts in visual form, and how does refining this skill, in turn, affect recognition? We developed an online platform for collecting large amounts of drawing and recognition data, and applied a deep convolutional neural network model of visual cortex trained only on natural images to explore the hypothesis that drawing recruits the same abstract feature representations that support natural visual object recognition. Consistent with this hypothesis, higher layers of this model captured the abstract features of both drawings and natural images most important for recognition, and people learning to produce more recognizable drawings of objects exhibited enhanced recognition of those objects. These findings could explain why drawing is so effective for communicating visual concepts, they suggest novel approaches for evaluating and refining conceptual knowledge, and they highlight the potential of deep networks for understanding human learning.

neuroscience↗

bioRxiv2016

Neural differentiation of incorrectly predicted memories

When an item is predicted in a particular context but the prediction is violated, memory for that item is weakened (Kim et al., 2014). Here we explore what happens when such previously mispredicted items are later re-encountered. According to prior neural network simulations, this sequence of events - misprediction and subsequent restudy - should lead to differentiation of the item's neural representation from the previous context (on which the misprediction was based). Specifically, misprediction weakens connections in the representation to features shared with the previous context, and restudy allows new features to be incorporated into the representation that are not shared with the previous context. This cycle of misprediction and restudy should have the net effect of moving the item s neural representation away from the neural representation of the previous context. We tested this hypothesis using fMRI, by tracking changes in item-specific BOLD activity patterns in the hippocampus, a key structure for representing memories and generating predictions. In left CA2/3/DG, we found greater neural differentiation for items that were repeatedly mispredicted and restudied compared to items from a control condition that was identical except without misprediction. We also measured prediction strength in a trial-by-trial fashion and found that greater misprediction for an item led to more differentiation, further supporting our hypothesis. Thus, the consequences of prediction error go beyond memory weakening: If the mispredicted item is restudied, the brain adaptively differentiates its memory representation to improve the accuracy of subsequent predictions and to shield it from further weakening.\n\nSignificanceCompetition between overlapping memories leads to weakening of non-target memories over time, making it easier to access target memories. However, a non-target memory in one context might become a target memory in another context. How do such memories get re-strengthened without increasing competition again? Computational models suggest that the brain handles this by reducing neural connections to the previous context and adding connections to new features that were not part of the previous context. The result is neural differentiation away from the previous context. Here provide support for this theory, using fMRI to track neural representations of individual memories in the hippocampus and how they change based on learning.

neuroscience↗