Correction: Data-driven analyses of motor impairments in animal models of neurological disorders.

[This corrects the article DOI: 10.1371/journal.pbio.3000516.].

Integration of valence and conflict processing through cellular-field interactions in human subgenual cingulate during emotional face processing in treatment-resistant depression.

The subgenual anterior cingulate cortex (sgACC) has been identified as a key brain area involved in various cognitive and emotional processes. While the sgACC has been implicated in both emotional valuation and emotional conflict monitoring, it is still unclear how this area integrates multiple functions. We characterized both single neuron and local field oscillatory activity in 14 patients undergoing sgACC deep brain stimulation for treatment-resistant depression. During recording, patients were presented with a modified Stroop task containing emotional face images that varied in valence and congruence. We further analyzed spike-field interactions to understand how network dynamics influence single neuron activity in this area. Most single neurons responded to both valence and congruence, revealing that sgACC neuronal activity can encode multiple processes within the same task, indicative of multifunctionality. During peak neuronal response, we observed increased spectral power in low frequency oscillations, including theta-band synchronization (4-8 Hz), as well as desynchronization in beta-band frequencies (13-30 Hz). Theta activity was modulated by current trial congruency with greater increases in spectral power following non-congruent stimuli, while beta desynchronizations occurred regardless of emotional valence. Spike-field interactions revealed that local sgACC spiking was phase-locked most prominently to the beta band, whereas phase-locking to the theta band occurred in fewer neurons overall but was modulated more strongly for neurons that were responsive to task. Our findings provide the first direct evidence of spike-field interactions relating to emotional cognitive processing in the human sgACC. Furthermore, we directly related theta oscillatory dynamics in human sgACC to current trial congruency, demonstrating it as an important regulator during conflict detection. Our data endorse the sgACC as an integrative hub for cognitive emotional processing through modulation of beta and theta network activity.

Non-trivial relationship between behavioral avalanches and internal neuronal dynamics in a recurrent neural network.

Neuronal activity gives rise to behavior, and behavior influences neuronal dynamics, in a closed-loop control system. Is it possible then, to find a relationship between the statistical properties of behavior and neuronal dynamics? Measurements of neuronal activity and behavior have suggested a direct relationship between scale-free neuronal and behavioral dynamics. Yet, these studies captured only local dynamics in brain sub-networks. Here, we investigate the relationship between internal dynamics and output statistics in a mathematical model system where we have access to the dynamics of all network units. We train a recurrent neural network (RNN), initialized in a high-dimensional chaotic state, to sustain behavioral states for durations following a power-law distribution as observed experimentally. Changes in network connectivity due to training affect the internal dynamics of neuronal firings, leading to neuronal avalanche size distributions approximating power-laws over some ranges. Yet, randomizing the changes in network connectivity can leave these power-law features largely unaltered. Specifically, whereas neuronal avalanche duration distributions show some variations between RNNs with trained and randomized decoders, neuronal avalanche size distributions are invariant, in the total population and in output-correlated sub-populations. This is true independent of whether the randomized decoders preserve power-law distributed behavioral dynamics. This demonstrates that a one-to-one correspondence between the considered statistical features of behavior and neuronal dynamics cannot be established and their relationship is non-trivial. Our findings also indicate that statistical properties of the intrinsic dynamics may be preserved, even as the internal state responsible for generating the desired output dynamics is perturbed.

Data-driven analyses of motor impairments in animal models of neurological disorders.

Behavior provides important insights into neuronal processes. For example, analysis of reaching movements can give a reliable indication of the degree of impairment in neurological disorders such as stroke, Parkinson disease, or Huntington disease. The analysis of such movement abnormalities is notoriously difficult and requires a trained evaluator. Here, we show that a deep neural network is able to score behavioral impairments with expert accuracy in rodent models of stroke. The same network was also trained to successfully score movements in a variety of other behavioral tasks. The neural network also uncovered novel movement alterations related to stroke, which had higher predictive power of stroke volume than the movement components defined by human experts. Moreover, when the regression network was trained only on categorical information (control = 0; stroke = 1), it generated predictions with intermediate values between 0 and 1 that matched the human expert scores of stroke severity. The network thus offers a new data-driven approach to automatically derive ratings of motor impairments. Altogether, this network can provide a reliable neurological assessment and can assist the design of behavioral indices to diagnose and monitor neurological disorders.

Lesions of ventrolateral striatum eliminate lose-shift but not win-stay behaviour in rats.

Animals tend to repeat actions that are associated with reward delivery, whereas they tend to shift responses to alternate choices following reward omission. These so-called win-stay and lose-shift responses are employed by a wide range of animals in a variety of decision-making scenarios, and depend on dissociated regions of the striatum. Specifically, lose-shift responding is impaired by extensive excitotoxic lesions of the lateral striatum. Here we used focal lesions to assess whether dorsal and ventral regions of the lateral striatum contribute differently to this effect. We found that damage to ventrolateral striatum reduced lose-shift responding without impairing win-stay, motoric, or motivational aspects of behaviour in the task, whereas lesions confined to the dorsolateral striatum significantly impaired the ability of rats to complete trials of the task. Moreover, lesions to the dorsomedial striatum had no effect on either lose-shift or win-stay responding. Together, these data suggest a novel role of the ventral portion of the lateral striatum in driving lose-shift decisions.

Implicit Valuation of the Near-Miss is Dependent on Outcome Context.

Gambling studies have described a "near-miss effect" wherein the experience of almost winning increases gambling persistence. The near-miss has been proposed to inflate the value of preceding actions through its perceptual similarity to wins. We demonstrate here, however, that it acts as a conditioned stimulus to positively or negatively influence valuation, dependent on reward expectation and cognitive engagement. When subjects are asked to choose between two simulated slot machines, near-misses increase valuation of machines with a low payout rate, whereas they decrease valuation of high payout machines. This contextual effect impairs decisions and persists regardless of manipulations to outcome feedback or financial incentive provided for good performance. It is consistent with proposals that near-misses cause frustration when wins are expected, and we propose that it increases choice stochasticity and overrides avoidance of low-valued options. Intriguingly, the near-miss effect disappears when subjects are required to explicitly value machines by placing bets, rather than choosing between them. We propose that this task increases cognitive engagement and recruits participation of brain regions involved in cognitive processing, causing inhibition of otherwise dominant systems of decision-making. Our results reveal that only implicit, rather than explicit strategies of decision-making are affected by near-misses, and that the brain can fluidly shift between these strategies according to task demands.

Lesions of dorsal striatum eliminate lose-switch responding but not mixed-response strategies in rats.

We used focal brain lesions in rats to examine how dorsomedial (DMS) and dorsolateral (DLS) regions of the striatum differently contribute to response adaptation driven by the delivery or omission of rewards. Rats performed a binary choice task under two modes: one in which responses were rewarded on half of the trials regardless of choice; and another 'competitive' one in which only unpredictable choices were rewarded. In both modes, control animals were more likely to use a predictable lose-switch strategy than animals with lesions of either DMS or DLS. Animals with lesions of DMS presumably relied more on DLS for behavioural control, and generated repetitive responses in the first mode. These animals then shifted to a random response strategy in the competitive mode, thereby performing better than controls or animals with DLS lesions. Analysis using computational models of reinforcement learning indicated that animals with striatal lesions, particularly of the DLS, had blunted reward sensitivity and less stochasticity in the choice mechanism. These results provide further evidence that the rodent DLS is involved in rapid response adaptation that is more sophisticated than that embodied by the classic notion of habit formation driven by gradual stimulus-response learning.

The Nonclassic Psychedelic Ibogaine Disrupts Cognitive Maps.

Background: The ability of psychedelic compounds to profoundly alter mental function has been long known, but the underlying changes in cellular-level information encoding remain poorly understood. Methods: We used two-photon microscopy to record from the retrosplenial cortex in head-fixed mice running on a treadmill before and after injection of the nonclassic psychedelic ibogaine (40 mg/kg intraperitoneally). Results: We found that the cognitive map, formed by the representation of position encoded by ensembles of individual neurons in the retrosplenial cortex, was destabilized by ibogaine when mice had to infer position between tactile landmarks. This corresponded with increased neural activity rates, loss of correlation structure, and increased responses to cues. Ibogaine had surprisingly little effect on the size-frequency distribution of network activity events, suggesting that signal propagation within the retrosplenial cortex was largely unaffected. Conclusions: Taken together, these data support proposals that compounds with psychedelic properties disrupt representations that are important for constraining neocortical activity, thereby increasing the entropy of neural signaling. Furthermore, the loss of expected position encoding between landmarks recapitulated effects of hippocampal impairment, suggesting that disruption of cognitive maps or other hippocampal processing may be a contributing mechanism of discoordinated neocortical activity in psychedelic states.

Changes in functional connectivity preserve scale-free neuronal and behavioral dynamics.

Does the brain optimize itself for storage and transmission of information, and if so, how? The critical brain hypothesis is based in statistical physics and posits that the brain self-tunes its dynamics to a critical point or regime to maximize the repertoire of neuronal responses. Yet, the robustness of this regime, especially with respect to changes in the functional connectivity, remains an unsolved fundamental challenge. Here, we show that both scale-free neuronal dynamics and self-similar features of behavioral dynamics persist following significant changes in functional connectivity. Specifically, we find that the psychedelic compound ibogaine that is associated with an altered state of consciousness fundamentally alters the functional connectivity in the retrosplenial cortex of mice. Yet, the scale-free statistics of movement and of neuronal avalanches among behaviorally related neurons remain largely unaltered. This indicates that the propagation of information within biological neural networks is robust to changes in functional organization of subpopulations of neurons, opening up a new perspective on how the adaptive nature of functional networks may lead to optimality of information transmission in the brain.

Cannabis use is associated with sexually dimorphic changes in executive control of visuospatial decision-making.

When the outcome of a choice is less favorable than expected, humans and animals typically shift to an alternate choice option on subsequent trials. Several lines of evidence indicate that this "lose-shift" responding is an innate sensorimotor response strategy that is normally suppressed by executive function. Therefore, the lose-shift response provides a covert gauge of cognitive control over choice mechanisms. We report here that the spatial position, rather than visual features, of choice targets drives the lose-shift effect. Furthermore, the ability to inhibit lose-shift responding to gain reward is different among male and female habitual cannabis users. Increased self-reported cannabis use was concordant with suppressed response flexibility and an increased tendency to lose-shift in women, which reduced performance in a choice task in which random responding is the optimal strategy. On the other hand, increased cannabis use in men was concordant with reduced reliance on spatial cues during decision-making, and had no impact on the number of correct responses. These data (63,600 trials from 106 participants) provide strong evidence that spatial-motor processing is an important component of economic decision-making, and that its governance by executive systems is different in men and women who use cannabis frequently.

Acute gut inflammation reduces neural activity and spine maturity in hippocampus but not basolateral amygdala.

Gastrointestinal tract (gut) inflammation increases stress and threat-coping behaviors, which are associated with altered activity in fear-related neural circuits, such as the basolateral amygdala and hippocampus. It remains to be determined whether inflammation from the gut affects neural activity by altering dendritic spines. We hypothesized that acute inflammation alters dendritic spines in a brain region-specific manner. Here we show that acute gut inflammation (colitis) evoked by dextran sodium sulfate (DSS) did not affect the overall spine density in the CA1 region of hippocampus, but increased the relative proportion of immature spines to mature spines on basal dendrites of pyramidal neurons. In contrast, in animals with colitis, no changes in spine density or composition on dendrites of pyramidal cells was observed in the basolateral amygdala. Rather, we observed decreased spine density on dendrites of stellate neurons, but not the relative proportions of mature vs immature spines. We used cFos expression evoked by the forced swim task as a measure of neural activity during stress and found no effect of DSS on the density of cFos immunoreactive neurons in basolateral amygdala. In contrast, fewer CA1 neurons expressed cFos in mice with colitis, relative to controls. Furthermore, CA1 cFos expression negatively correlated with active stress-coping in the swim task and was negatively correlated with gut inflammation. These data reveal that the effects of acute gut inflammation on synaptic remodeling depend on brain region, neuronal phenotype, and dendrite location. In the hippocampus, a shift to immature spines and hypoactivity are more strongly related to colitis-evoked behavioral changes than is remodeling in basolateral amygdala.

More is less: a disinhibited prefrontal cortex impairs cognitive flexibility.

The prefrontal cortex (PFC) is critical for decision making, and it becomes dysfunctional in many neuropsychiatric disorders. Studies in schizophrenia patients and relevant animal models suggest loss of PFC inhibitory interneuron function. For instance, rats with a neonatal ventral hippocampal lesion (NVHL) show a deficient modulation of PFC interneurons by dopamine (DA). Whether the PFC becomes disinhibited in this model and alters decision making remains to be determined. Here, we recorded neural activity in the medial PFC of NVHL rats during a reward-discounting choice task that activated DA systems. Rats were trained to sample odors that instructed them to select one of two feeders that delivered unequal amounts of liquid. Putative pyramidal neurons in the PFC were hyperactive whereas task-related field potential oscillations were significantly reduced in NVHL rats, consistent with impaired interneuron activation by DA during odor sampling leading to disorganized processing. Cognitive flexibility was tested by examining response bias and errors after reversing reward outcomes. NVHL rats demonstrated impaired flexibility as they were less able to track changes in reward outcome and made more response errors than controls did. Reducing cortical excitability with the metabotropic glutamate receptor 2/3 agonist LY379268 (1 mg/kg, i.p.) improved behavioral flexibility in NVHL rats but not controls. Furthermore, D2 dopamine receptors were involved, as the antagonist eticlopride (0.02 mg/kg, i.p.) reduced the ability to switch only in control animals. We conclude that NVHL rats present PFC disinhibition, which affects neural information processing and the selection of appropriate behavioral responses.

Amphetamine reduces reward encoding and stabilizes neural dynamics in rat anterior cingulate cortex.

Psychostimulants such as d-amphetamine (AMPH) often have behavioral effects that appear paradoxical within the framework of optimal choice theory. AMPH typically increases task engagement and the effort animals exert for reward, despite decreasing reward valuation. We investigated neural correlates of this phenomenon in the anterior cingulate cortex (ACC), a brain structure implicated in signaling cost-benefit utility. AMPH decreased signaling of reward, but not effort, in the ACC of freely-moving rats. Ensembles of simultaneously recorded neurons generated task-specific trajectories of neural activity encoding past, present, and future events. Low-dose AMPH contracted these trajectories and reduced their variance, whereas high-dose AMPH expanded both. We propose that under low-dose AMPH, increased network stability balances moderately increased excitability, which promotes accelerated unfolding of a neural 'script' for task execution, despite reduced reward valuation. Noise from excessive excitability at high doses overcomes stability enhancement to drive frequent deviation from the script, impairing task execution.

Behavioral adaptations in a relapsing mouse model of colitis.

Inflammatory bowel disease (IBD) is characterized by relapsing periods of gut inflammation, and is comorbid with depression, anxiety, and cognitive deficits. Animal models of IBD that explore the behavioral consequences almost exclusively use acute models of gut inflammation, which fails to recapitulate the cyclic, chronic nature of IBD. This study sought to identify behavioral differences in digging, memory, and stress-coping strategies in mice exposed to one (acute) or three (chronic) cycles of gut inflammation, using the dextran sodium sulfate (DSS) model of colitis. Similar levels of gut pathology were observed between acute and chronically exposed mice, although mice in the chronic treatment had significantly shorter colons, suggesting more severe disease. Behavioral measures revealed an unexpected pattern in which chronic treatment evoked fewer deficits than acute treatment. Specifically, acutely-treated mice showed alterations in measures of object burying, novel object recognition, object location memory, and stress-coping (forced swim task). Chronically-treated animals, however, showed similar alterations in object burying, but not the other measures. These data suggest an adaptive or tolerizing effect of repeated cycles of peripheral gut inflammation on mnemonic function and stress-coping, whereas some other behaviors continue to be affected by gut inflammation. We speculate that the normalization of some functions may involve the reversion to the baseline state of the hypothalamic-pituitary-adrenal axis and/or levels of neuroinflammation, which are both activated by the first exposure to the colitic agent.

Bursting activation of prefrontal cortex drives sustained up states in nucleus accumbens spiny neurons in vivo.

Hippocampal inputs to the nucleus accumbens (NA) have been proposed to implement a gating mechanism by driving NA medium spiny neurons (MSNs) to depolarized up states that facilitate action potential firing in response to brief activation of the prefrontal cortex (PFC). Brief PFC stimulation alone, on the other hand, could not drive NA up states. As these studies were conducted using single-pulse PFC stimulation, it remains possible that PFC activation with naturalistic, bursty patterns can also drive up states in NA MSNs. Here, we assessed NA responses to PFC stimulation with a pattern similar to what is typically observed in awake animals during PFC-relevant behaviors. In vivo intracellular recordings from NA MSNs revealed that brief 20-50 Hz PFC stimulus trains evoked depolarizations that were similar to spontaneous up states in NA MSNs and were sustained beyond stimulus offset. Similar train stimulation of corticoaccumbens afferents in a parasagittal slice preparation evoked large amplitude depolarizations in NA MSNs that were sustained during stimulation but decayed rapidly following stimulation offset, suggesting that activation of cortical afferents can drive MSN depolarizations but other mechanisms may contribute to sustaining up states. These data suggest that NA MSNs integrate temporal features of PFC activation and that the NA gating model can be reformulated to include a PFC-driven gating mechanism during periods of high PFC firing, such as during cognitively demanding tasks. Synapse 63:173-180, 2009. (c) 2008 Wiley-Liss, Inc.

Phase of electroencephalography theta oscillation during stimulus encoding affects accuracy of memory recall.

Oscillatory activity is a ubiquitous property of brain signals, and yet relatively few studies have investigated how the phase of such ongoing oscillations affects our cognition. One of the main findings in this field is that the phase of electroencephalography (EEG) in the alpha band can affect perception of milliseconds-long stimuli. However, the importance of the phase of EEG for processing more naturalistic stimuli, which have a much longer duration, is still not clear. To address this question here, we presented word-nonword pairs, each of which was visible for 5 s and measured the effect of EEG phase during stimulus onset on later memory recall. The task consisted of an encoding (learning) phase in which 20 novel word-nonword pairs were presented, followed by a test phase in which participants were shown one of the seen words with four target nonwords to choose from. We found that memory recall performance was higher when the words during encoding were presented at a descending phase of the theta oscillation. This effect was the strongest in the frontal cortex. These results suggest that the phase of ongoing cortical activity can affect memorization of seconds-long stimuli that are an integral part of many daily tasks.

Distributed Encoding of Reinforcement in Rat Cortico-Striatal-Limbic Networks.

Decision-making in the mammalian brain typically involves multiple brain structures within the midbrain, thalamus, striatum, limbic system, and cortex. Although task specific contributions of each brain region have been identified, neurons responding to reinforcement have been found throughout these structures. We sought to determine if any brain area, or cluster of areas, are the source of information, and if the fidelity of information varies among the areas. We recorded simultaneous field potentials (FPs) in rats from seven brain regions as they completed a binary choice task. The FPs of a 0.5 s window following reinforcement were given as input to a classifier that attempted to predict whether or not the rat received reward on each trial. The classifier correctly categorized reward on 77% of trials. Any region-specific signal could be omitted without lowering accuracy. Frequencies above 40 Hz and signals recorded later than 0.25 s following reinforcement were necessary to achieve this accuracy. Further, the classifier was able to predict reinforcement outcome above chance levels when using FPs from any single recorded brain region. Some combinations of structures, however, were more predictive than others. Analysis of FPs prior to reward revealed most regions reflected the prior probability of reward. Lastly, analyses of information flow suggested reinforcement information does not originate within a single structure of the network, within the resolution afforded by FP recordings. These data suggest reward delivery information is rapidly distributed non-uniformly across the network, and there is no canonical flow of information about reward events in the recorded structures.

Using Neuron Spiking Activity to Trigger Closed-Loop Stimuli in Neurophysiological Experiments.

Closed-loop neurophysiological systems use patterns of neuronal activity to trigger stimuli, which in turn affect brain activity. Such closed-loop systems are already found in clinical applications, and are important tools for basic brain research. A particularly interesting recent development is the integration of closed-loop approaches with optogenetics, such that specific patterns of neuronal activity can trigger optical stimulation of selected neuronal groups. However, setting up an electrophysiological system for closed-loop experiments can be difficult. Here, a ready-to-apply Matlab code is provided for triggering stimuli based on the activity of single or multiple neurons. This sample code can be easily modified based on individual needs. For instance, it shows how to trigger sound stimuli and how to change it to trigger an external device connected to a PC serial port. The presented protocol is designed to work with a popular neuronal recording system for animal studies (Neuralynx). The implementation of closed-loop stimulation is demonstrated in an awake rat.

Lesions of lateral habenula attenuate win-stay but not lose-shift responses in a competitive choice task.

Multiple neural systems contribute to choice adaptation following reinforcement. Recent evidence suggests that the lateral habenula (LHb) plays a key role in such adaptations, particularly when reinforcements are worse than expected. Here, we investigated the effects of bilateral LHb lesions on responding in a binary choice task with no discriminatory cues. LHb lesions in rats decreased win-stay responses but surprisingly left lose-shift responses intact. This same dissociated effect was also observed after systemic administration of d-amphetamine in a separate cohort of animals. These results suggest that at least some behavioural responses triggered by reward omission do not depend on an intact LHb.

Learning to Predict Consequences as a Method of Knowledge Transfer in Reinforcement Learning.

The reinforcement learning (RL) paradigm allows agents to solve tasks through trial-and-error learning. To be capable of efficient, long-term learning, RL agents should be able to apply knowledge gained in the past to new tasks they may encounter in the future. The ability to predict actions' consequences may facilitate such knowledge transfer. We consider here domains where an RL agent has access to two kinds of information: agent-centric information with constant semantics across tasks, and environment-centric information, which is necessary to solve the task, but with semantics that differ between tasks. For example, in robot navigation, environment-centric information may include the robot's geographic location, while agent-centric information may include sensor readings of various nearby obstacles. We propose that these situations provide an opportunity for a very natural style of knowledge transfer, in which the agent learns to predict actions' environmental consequences using agent-centric information. These predictions contain important information about the affordances and dangers present in a novel environment, and can effectively transfer knowledge from agent-centric to environment-centric learning systems. Using several example problems including spatial navigation and network routing, we show that our knowledge transfer approach can allow faster and lower cost learning than existing alternatives.