Lightning Pose: improved animal pose estimation via semi-supervised learning, Bayesian ensembling and cloud-native open-source tools.

Contemporary pose estimation methods enable precise measurements of behavior via supervised deep learning with hand-labeled video frames. Although effective in many cases, the supervised approach requires extensive labeling and often produces outputs that are unreliable for downstream analyses. Here, we introduce 'Lightning Pose', an efficient pose estimation package with three algorithmic contributions. First, in addition to training on a few labeled video frames, we use many unlabeled videos and penalize the network whenever its predictions violate motion continuity, multiple-view geometry and posture plausibility (semi-supervised learning). Second, we introduce a network architecture that resolves occlusions by predicting pose on any given frame using surrounding unlabeled frames. Third, we refine the pose predictions post hoc by combining ensembling and Kalman smoothing. Together, these components render pose trajectories more accurate and scientifically usable. We released a cloud application that allows users to label data, train networks and process new videos directly from the browser.

Partitioning variability in animal behavioral videos using semi-supervised variational autoencoders.

Recent neuroscience studies demonstrate that a deeper understanding of brain function requires a deeper understanding of behavior. Detailed behavioral measurements are now often collected using video cameras, resulting in an increased need for computer vision algorithms that extract useful information from video data. Here we introduce a new video analysis tool that combines the output of supervised pose estimation algorithms (e.g. DeepLabCut) with unsupervised dimensionality reduction methods to produce interpretable, low-dimensional representations of behavioral videos that extract more information than pose estimates alone. We demonstrate this tool by extracting interpretable behavioral features from videos of three different head-fixed mouse preparations, as well as a freely moving mouse in an open field arena, and show how these interpretable features can facilitate downstream behavioral and neural analyses. We also show how the behavioral features produced by our model improve the precision and interpretation of these downstream analyses compared to using the outputs of either fully supervised or fully unsupervised methods alone.

Choices change the temporal weighting of decision evidence.

Many decisions result from the accumulation of decision-relevant information (evidence) over time. Even when maximizing decision accuracy requires weighting all the evidence equally, decision-makers often give stronger weight to evidence occurring early or late in the evidence stream. Here, we show changes in such temporal biases within participants as a function of intermittent judgments about parts of the evidence stream. Human participants performed a decision task that required a continuous estimation of the mean evidence at the end of the stream. The evidence was either perceptual (noisy random dot motion) or symbolic (variable sequences of numbers). Participants also reported a categorical judgment of the preceding evidence half-way through the stream in one condition or executed an evidence-independent motor response in another condition. The relative impact of early versus late evidence on the final estimation flipped between these two conditions. In particular, participants' sensitivity to late evidence after the intermittent judgment, but not the simple motor response, was decreased. Both the intermittent response as well as the final estimation reports were accompanied by nonluminance-mediated increases of pupil diameter. These pupil dilations were bigger during intermittent judgments than simple motor responses and bigger during estimation when the late evidence was consistent than inconsistent with the initial judgment. In sum, decisions activate pupil-linked arousal systems and alter the temporal weighting of decision evidence. Our results are consistent with the idea that categorical choices in the face of uncertainty induce a change in the state of the neural circuits underlying decision-making.NEW & NOTEWORTHY The psychology and neuroscience of decision-making have extensively studied the accumulation of decision-relevant information toward a categorical choice. Much fewer studies have assessed the impact of a choice on the processing of subsequent information. Here, we show that intermittent choices during a protracted stream of input reduce the sensitivity to subsequent decision information and transiently boost arousal. Choices might trigger a state change in the neural machinery for decision-making.

Adaptive History Biases Result from Confidence-Weighted Accumulation of past Choices.

Perceptual decision-making is biased by previous events, including the history of preceding choices: observers tend to repeat (or alternate) their judgments of the sensory environment more often than expected by chance. Computational models postulate that these so-called choice history biases result from the accumulation of internal decision signals across trials. Here, we provide psychophysical evidence for such a mechanism and its adaptive utility. Male and female human observers performed different variants of a challenging visual motion discrimination task near psychophysical threshold. In a first experiment, we decoupled categorical perceptual choices and motor responses on a trial-by-trial basis. Choice history bias was explained by previous perceptual choices, not motor responses, highlighting the importance of internal decision signals in action-independent formats. In a second experiment, observers performed the task in stimulus environments containing different levels of autocorrelation and providing no external feedback about choice correctness. Despite performing under overall high levels of uncertainty, observers adjusted both the strength and the sign of their choice history biases to these environments. When stimulus sequences were dominated by either repetitions or alternations, the individual degree of this adjustment of history bias was about as good a predictor of individual performance as individual perceptual sensitivity. The history bias adjustment scaled with two proxies for observers' confidence about their previous choices (accuracy and reaction time). Together, our results are consistent with the idea that action-independent, confidence-modulated decision variables are accumulated across choices in a flexible manner that depends on decision-makers' model of their environment.SIGNIFICANCE STATEMENT Decisions based on sensory input are often influenced by the history of one's preceding choices, manifesting as a bias to systematically repeat (or alternate) choices. We here provide support for the idea that such choice history biases arise from the context-dependent accumulation of a quantity referred to as the decision variable: the variable's sign dictates the choice and its magnitude the confidence about choice correctness. We show that choices are accumulated in an action-independent format and a context-dependent manner, weighted by the confidence about their correctness. This confidence-weighted accumulation of choices enables decision-makers to flexibly adjust their behavior to different sensory environments. The bias adjustment can be as important for optimizing performance as one's sensitivity to the momentary sensory input.

Reading memory formation from the eyes.

At any time, we are processing thousands of stimuli, but only few of them will be remembered hours or days later. Is there any way to predict which ones? Here, we tested whether the pupil response to ongoing stimuli, an indicator of physiological arousal known to be relevant for memory formation, is a reliable predictor of long-term memory for these stimuli, over at least 1 day. Pupil dilation was tracked while participants performed visual and auditory encoding tasks. Memory was tested immediately after encoding and 24 hr later. Irrespective of the encoding modality, trial-by-trial variations in pupil dilation predicted reliably which stimuli were recalled in the immediate and 24 hr-delayed tests, in particular for emotionally arousing stimuli. These results show that our eyes may provide a window into the formation of long-term memories. Furthermore, our findings underline the important role of central arousal systems in the rapid formation of memories in the brain, possibly by gating synaptic plasticity mechanisms in the neocortex.

A Bayesian Hierarchical Model of Trial-To-Trial Fluctuations in Decision Criterion.

Classical decision models assume that the parameters giving rise to choice behavior are stable, yet emerging research suggests these parameters may fluctuate over time. Such fluctuations, observed in neural activity and behavioral strategies, have significant implications for understanding decision-making processes. However, empirical studies on fluctuating human decision-making strategies have been limited due to the extensive data requirements for estimating these fluctuations. Here, we introduce hMFC (Hierarchical Model for Fluctuations in Criterion), a Bayesian framework designed to estimate slow fluctuations in the decision criterion from limited data. We first showcase the importance of considering fluctuations in decision criterion: incorrectly assuming a stable criterion gives rise to apparent history effects and underestimates perceptual sensitivity. We then present a hierarchical estimation procedure capable of reliably recovering the underlying state of the fluctuating decision criterion with as few as 500 trials per participant, offering a robust tool for researchers with typical human datasets. Critically, hMFC does not only accurately recover the state of the underlying decision criterion, it also effectively deals with the confounds caused by criterion fluctuations. Lastly, we provide code and a comprehensive demo at www.github.com/robinvloeberghs/hMFC to enable widespread application of hMFC in decision-making research.

Lightning Pose: improved animal pose estimation via semi-supervised learning, Bayesian ensembling, and cloud-native open-source tools.

Contemporary pose estimation methods enable precise measurements of behavior via supervised deep learning with hand-labeled video frames. Although effective in many cases, the supervised approach requires extensive labeling and often produces outputs that are unreliable for downstream analyses. Here, we introduce "Lightning Pose," an efficient pose estimation package with three algorithmic contributions. First, in addition to training on a few labeled video frames, we use many unlabeled videos and penalize the network whenever its predictions violate motion continuity, multiple-view geometry, and posture plausibility (semi-supervised learning). Second, we introduce a network architecture that resolves occlusions by predicting pose on any given frame using surrounding unlabeled frames. Third, we refine the pose predictions post-hoc by combining ensembling and Kalman smoothing. Together, these components render pose trajectories more accurate and scientifically usable. We release a cloud application that allows users to label data, train networks, and predict new videos directly from the browser.

Interactive data exploration websites for large-scale electrophysiology.

Methodological advances in neuroscience have enabled the collection of massive datasets which demand innovative approaches for scientific communication. Existing platforms for data storage lack intuitive tools for data exploration, limiting our ability to interact effectively with these brain-wide datasets. We introduce two public websites: (Data and Atlas) developed for the International Brain Laboratory which provide access to millions of behavioral trials and hundreds of thousands of individual neurons. These interfaces allow users to discover both the raw and processed brain-wide data released by the IBL at the scale of the whole brain, individual sessions, trials, and neurons. By hosting these data interfaces as websites they are available cross-platform with no installation. By releasing each site's code as a modular open-source framework, other researchers can easily develop their own web interfaces and explore their own data. As neuroscience datasets continue to expand, customizable web interfaces offer a glimpse into a future of streamlined data exploration and act as blueprints for future tools.

Rethinking academia in a time of climate crisis.

Addressing the climate crisis requires radical and urgent action at all levels of society. Universities are ideally positioned to lead such action but are largely failing to do so. At the same time, many academic scientists find their work impeded by bureaucracy, excessive competitiveness, and a loss of academic freedom. Here, drawing on the framework of "Doughnut Economics," developed by Kate Raworth, we suggest seven new principles for rethinking the norms of scientific practice. Based on these, we propose a call to action, and encourage academics to take concrete steps towards the creation of a flourishing scientific enterprise that is fit for the challenges of the 21st century.

Persistent activity in human parietal cortex mediates perceptual choice repetition bias.

Humans and other animals tend to repeat or alternate their previous choices, even when judging sensory stimuli presented in a random sequence. It is unclear if and how sensory, associative, and motor cortical circuits produce these idiosyncratic behavioral biases. Here, we combined behavioral modeling of a visual perceptual decision with magnetoencephalographic (MEG) analyses of neural dynamics, across multiple regions of the human cerebral cortex. We identified distinct history-dependent neural signals in motor and posterior parietal cortex. Gamma-band activity in parietal cortex tracked previous choices in a sustained fashion, and biased evidence accumulation toward choice repetition; sustained beta-band activity in motor cortex inversely reflected the previous motor action, and biased the accumulation starting point toward alternation. The parietal, not motor, signal mediated the impact of previous on current choice and reflected individual differences in choice repetition. In sum, parietal cortical signals seem to play a key role in shaping choice sequences.

Large-scale neural recordings call for new insights to link brain and behavior.

Neuroscientists today can measure activity from more neurons than ever before, and are facing the challenge of connecting these brain-wide neural recordings to computation and behavior. In the present review, we first describe emerging tools and technologies being used to probe large-scale brain activity and new approaches to characterize behavior in the context of such measurements. We next highlight insights obtained from large-scale neural recordings in diverse model systems, and argue that some of these pose a challenge to traditional theoretical frameworks. Finally, we elaborate on existing modeling frameworks to interpret these data, and argue that the interpretation of brain-wide neural recordings calls for new theoretical approaches that may depend on the desired level of understanding. These advances in both neural recordings and theory development will pave the way for critical advances in our understanding of the brain.

Climate crisis and ecological emergency: Why they concern (neuro)scientists, and what we can do.

Our planet is experiencing severe and accelerating climate and ecological breakdown caused by human activity. As professional scientists, we are better placed than most to understand the data that evidence this fact. However, like most other people, we ignore this inconvenient truth and lead our daily lives, at home and at work, as if these facts weren't true. In particular, we overlook that our own neuroscientific research practices, from our laboratory experiments to our often global travel, help drive climate change and ecosystem damage. We also hold privileged positions of authority in our societies but rarely speak out. Here, we argue that to help society create a survivable future, we neuroscientists can and must play our part. In April 2021, we delivered a symposium at the British Neuroscience Association meeting outlining what we think neuroscientists can and should do to help stop climate breakdown. Building on our talks (Box 1), we here outline what the climate and ecological emergencies mean for us as neuroscientists. We highlight the psychological mechanisms that block us from taking action, and then outline what practical steps we can take to overcome these blocks and work towards sustainability. In particular, we review environmental issues in neuroscience research, scientific computing, and conferences. We also highlight the key advocacy roles we can all play in our institutions and in society more broadly. The need for sustainable change has never been more urgent, and we call on all (neuro)scientists to act with the utmost urgency.

Mice alternate between discrete strategies during perceptual decision-making.

Classical models of perceptual decision-making assume that subjects use a single, consistent strategy to form decisions, or that decision-making strategies evolve slowly over time. Here we present new analyses suggesting that this common view is incorrect. We analyzed data from mouse and human decision-making experiments and found that choice behavior relies on an interplay among multiple interleaved strategies. These strategies, characterized by states in a hidden Markov model, persist for tens to hundreds of trials before switching, and often switch multiple times within a session. The identified decision-making strategies were highly consistent across mice and comprised a single 'engaged' state, in which decisions relied heavily on the sensory stimulus, and several biased states in which errors frequently occurred. These results provide a powerful alternate explanation for 'lapses' often observed in rodent behavioral experiments, and suggest that standard measures of performance mask the presence of major changes in strategy across trials.

Citric Acid Water as an Alternative to Water Restriction for High-Yield Mouse Behavior.

Powerful neural measurement and perturbation tools have positioned mice as an ideal species for probing the neural circuit mechanisms of cognition. Crucial to this success is the ability to motivate animals to perform specific behaviors. One successful strategy is to restrict their water intake, rewarding them with water during a behavioral task. However, water restriction requires rigorous monitoring of animals' health and hydration status and can be challenging for some mice. We present an alternative that allows mice more control over their water intake: free home-cage access to water, made slightly sour by a small amount of citric acid (CA). In a previous study, rats with free access to CA water readily performed a behavioral task for water rewards, although completing fewer trials than under water restriction (Reinagel, 2018). We here extend this approach to mice and confirm its robustness across multiple laboratories. Mice reduced their intake of CA water while maintaining healthy weights. Continuous home-cage access to CA water only subtly impacted their willingness to perform a decision-making task, in which they were rewarded with sweetened water. When free CA water was used instead of water restriction only on weekends, learning and decision-making behavior were unaffected. CA water is thus a promising alternative to water restriction, allowing animals more control over their water intake without interfering with behavioral performance.

Standardized and reproducible measurement of decision-making in mice.

Progress in science requires standardized assays whose results can be readily shared, compared, and reproduced across laboratories. Reproducibility, however, has been a concern in neuroscience, particularly for measurements of mouse behavior. Here, we show that a standardized task to probe decision-making in mice produces reproducible results across multiple laboratories. We adopted a task for head-fixed mice that assays perceptual and value-based decision making, and we standardized training protocol and experimental hardware, software, and procedures. We trained 140 mice across seven laboratories in three countries, and we collected 5 million mouse choices into a publicly available database. Learning speed was variable across mice and laboratories, but once training was complete there were no significant differences in behavior across laboratories. Mice in different laboratories adopted similar reliance on visual stimuli, on past successes and failures, and on estimates of stimulus prior probability to guide their choices. These results reveal that a complex mouse behavior can be reproduced across multiple laboratories. They establish a standard for reproducible rodent behavior, and provide an unprecedented dataset and open-access tools to study decision-making in mice. More generally, they indicate a path toward achieving reproducibility in neuroscience through collaborative open-science approaches.

In science, it is of vital importance that multiple studies corroborate the same result. Researchers therefore need to know all the details of previous experiments in order to implement the procedures as exactly as possible. However, this is becoming a major problem in neuroscience, as animal studies of behavior have proven to be hard to reproduce, and most experiments are never replicated by other laboratories. Mice are increasingly being used to study the neural mechanisms of decision making, taking advantage of the genetic, imaging and physiological tools that are available for mouse brains. Yet, the lack of standardized behavioral assays is leading to inconsistent results between laboratories. This makes it challenging to carry out large-scale collaborations which have led to massive breakthroughs in other fields such as physics and genetics. To help make these studies more reproducible, the International Brain Laboratory (a collaborative research group) et al. developed a standardized approach for investigating decision making in mice that incorporates every step of the process; from the training protocol to the software used to analyze the data. In the experiment, mice were shown images with different contrast and had to indicate, using a steering wheel, whether it appeared on their right or left. The mice then received a drop of sugar water for every correction decision. When the image contrast was high, mice could rely on their vision. However, when the image contrast was very low or zero, they needed to consider the information of previous trials and choose the side that had recently appeared more frequently. This method was used to train 140 mice in seven laboratories from three different countries. The results showed that learning speed was different across mice and laboratories, but once training was complete the mice behaved consistently, relying on visual stimuli or experiences to guide their choices in a similar way. These results show that complex behaviors in mice can be reproduced across multiple laboratories, providing an unprecedented dataset and open-access tools for studying decision making. This work could serve as a foundation for other groups, paving the way to a more collaborative approach in the field of neuroscience that could help to tackle complex research challenges.

How Can Neuroscientists Respond to the Climate Emergency?

The world faces a climate emergency. Here, we consider the actions that can be taken by neuroscientists to tackle climate change. We encourage neuroscientists to put emissions reductions at the center of their everyday professional activities.

Pupil-linked phasic arousal predicts a reduction of choice bias across species and decision domains.

Decisions are often made by accumulating ambiguous evidence over time. The brain's arousal systems are activated during such decisions. In previous work in humans, we found that evoked responses of arousal systems during decisions are reported by rapid dilations of the pupil and track a suppression of biases in the accumulation of decision-relevant evidence (de Gee et al., 2017). Here, we show that this arousal-related suppression in decision bias acts on both conservative and liberal biases, and generalizes from humans to mice, and from perceptual to memory-based decisions. In challenging sound-detection tasks, the impact of spontaneous or experimentally induced choice biases was reduced under high phasic arousal. Similar bias suppression occurred when evidence was drawn from memory. All of these behavioral effects were explained by reduced evidence accumulation biases. Our results point to a general principle of interplay between phasic arousal and decision-making.

Reinforcement biases subsequent perceptual decisions when confidence is low, a widespread behavioral phenomenon.

Learning from successes and failures often improves the quality of subsequent decisions. Past outcomes, however, should not influence purely perceptual decisions after task acquisition is complete since these are designed so that only sensory evidence determines the correct choice. Yet, numerous studies report that outcomes can bias perceptual decisions, causing spurious changes in choice behavior without improving accuracy. Here we show that the effects of reward on perceptual decisions are principled: past rewards bias future choices specifically when previous choice was difficult and hence decision confidence was low. We identified this phenomenon in six datasets from four laboratories, across mice, rats, and humans, and sensory modalities from olfaction and audition to vision. We show that this choice-updating strategy can be explained by reinforcement learning models incorporating statistical decision confidence into their teaching signals. Thus, reinforcement learning mechanisms are continually engaged to produce systematic adjustments of choices even in well-learned perceptual decisions in order to optimize behavior in an uncertain world.

Harnessing behavioral diversity to understand neural computations for cognition.

With the increasing acquisition of large-scale neural recordings comes the challenge of inferring the computations they perform and understanding how these give rise to behavior. Here, we review emerging conceptual and technological advances that begin to address this challenge, garnering insights from both biological and artificial neural networks. We argue that neural data should be recorded during rich behavioral tasks, to model cognitive processes and estimate latent behavioral variables. Careful quantification of animal movements can also provide a more complete picture of how movements shape neural dynamics and reflect changes in brain state, such as arousal or stress. Artificial neural networks (ANNs) could serve as artificial model organisms to connect neural dynamics and rich behavioral data. ANNs have already begun to reveal how a wide range of different behaviors can be implemented, generating hypotheses about how observed neural activity might drive behavior and explaining diversity in behavioral strategies.