Recurrent Network Dynamics; a Link between Form and Motion.

To discriminate visual features such as corners and contours, the brain must be sensitive to spatial correlations between multiple points in an image. Consistent with this, macaque V2 neurons respond selectively to patterns with well-defined multipoint correlations. Here, we show that a standard feedforward model (a cascade of linear-non-linear filters) does not capture this multipoint selectivity. As an alternative, we developed an artificial neural network model with two hierarchical stages of processing and locally recurrent connectivity. This model faithfully reproduced neurons' selectivity for multipoint correlations. By probing the model, we gained novel insights into early form processing. First, the diverse selectivity for multipoint correlations and complex response dynamics of the hidden units in the model were surprisingly similar to those observed in V1 and V2. This suggests that both transient and sustained response dynamics may be a vital part of form computations. Second, the model self-organized units with speed and direction selectivity that was correlated with selectivity for multipoint correlations. In other words, the model units that detected multipoint spatial correlations also detected space-time correlations. This leads to the novel hypothesis that higher-order spatial correlations could be computed by the rapid, sequential assessment and comparison of multiple low-order correlations within the receptive field. This computation links spatial and temporal processing and leads to the testable prediction that the analysis of complex form and motion are closely intertwined in early visual cortex.

Visual processing of informative multipoint correlations arises primarily in V2.

Using the visual system as a model, we recently showed that the efficient coding principle accounted for the allocation of computational resources in central sensory processing: when sampling an image is the main limitation, resources are devoted to compute the statistical features that are the most variable, and therefore the most informative (eLife 2014;3:e03722. DOI: 10.7554/eLife.03722 Hermundstad et al., 2014). Building on these results, we use single-unit recordings in the macaque monkey to determine where these computations--sensitivity to specific multipoint correlations--occur. We find that these computations take place in visual area V2, primarily in its supragranular layers. The demonstration that V2 neurons are sensitive to the multipoint correlations that are informative about natural images provides a common computational underpinning for diverse but well-recognized aspects of neural processing in V2, including its sensitivity to corners, junctions, illusory contours, figure/ground, and 'naturalness.'

Estimating the amount of information conveyed by a population of neurons.

Recent technological advances have made the simultaneous recording of the activity of many neurons common. However, estimating the amount of information conveyed by the discharge of a neural population remains a significant challenge. Here we describe our recently published analysis method that assists in such estimates. We describe the key concepts and assumptions on which the method is based, illustrate its use with data from both simulated and real neurons recorded from the lateral geniculate nucleus of a monkey, and show how it can be used to calculate redundancy and synergy among neuronal groups.

Estimating the amount of information carried by a neuronal population.

Although all brain functions require coordinated activity of many neurons, it has been difficult to estimate the amount of information carried by a population of spiking neurons. We present here a Fourier-based method for estimating the information delivery rate from a population of neurons, which allows us to measure the redundancy of information within and between functional neuronal classes. We illustrate the use of the method on some artificial spike trains and on simultaneous recordings from a small population of neurons from the lateral geniculate nucleus of an anesthetized macaque monkey.

Homeostasis of exercise hyperpnea and optimal sensorimotor integration: the internal model paradigm.

Homeostasis is a basic tenet of biomedicine and an open problem for many physiological control systems. Among them, none has been more extensively studied and intensely debated than the dilemma of exercise hyperpnea - a paradoxical homeostatic increase of respiratory ventilation that is geared to metabolic demands instead of the normal chemoreflex mechanism. Classical control theory has led to a plethora of "feedback/feedforward control" or "set point" hypotheses for homeostatic regulation, yet so far none of them has proved satisfactory in explaining exercise hyperpnea and its interactions with other respiratory inputs. Instead, the available evidence points to a far more sophisticated respiratory controller capable of integrating multiple afferent and efferent signals in adapting the ventilatory pattern toward optimality relative to conflicting homeostatic, energetic and other objectives. This optimality principle parsimoniously mimics exercise hyperpnea, chemoreflex and a host of characteristic respiratory responses to abnormal gas exchange or mechanical loading/unloading in health and in cardiopulmonary diseases - all without resorting to a feedforward "exercise stimulus". Rather, an emergent controller signal encoding the projected metabolic level is predicted by the principle as an exercise-induced 'mental percept' or 'internal model', presumably engendered by associative learning (operant conditioning or classical conditioning) which achieves optimality through continuous identification of, and adaptation to, the causal relationship between respiratory motor output and resultant chemical-mechanical afferent feedbacks. This internal model self-tuning adaptive control paradigm opens a new challenge and exciting opportunity for experimental and theoretical elucidations of the mechanisms of respiratory control - and of homeostatic regulation and sensorimotor integration in general.

Cytoarchitecture of pneumotaxic integration of respiratory and nonrespiratory information in the rat.

The "pneumotaxic center" in the Kölliker-Fuse and medial parabrachial nuclei of dorsolateral pons (dl-pons) plays an important role in respiratory phase switching, modulation of respiratory reflex, and rhythmogenesis. Recent electrophysiological and neural tracing data implicate additional pneumotaxic nuclei in (and a broader role for) the dl-pons in integrating respiratory and nonrespiratory information. Here, we examined the cytoarchitecture of the greater pneumotaxic center and its integrating function by using combined extracellular recording and juxtacellular labeling of unit respiratory rhythmic neurons in dl-pons in urethane-anesthetized, vagotomized, paralyzed, and servo-ventilated adult Sprague Dawley rats. Perievent histogram analysis identified four major types of neuronal discharge patterns: inspiratory, expiratory (with three subdivisions), inspiratory-expiratory, and expiratory-inspiratory phase spanning, sometimes with mild tonic background activity. Most recorded neurons were localized in the Kölliker-Fuse and medial parabrachial nuclei, but some were also found in lateral parabrachial nucleus, intertrigeminal nucleus, principal trigeminal sensory nucleus, and supratrigeminal nucleus. The majority of labeled neurons had large and spatially extended dendritic trees that spanned several of these dl-pons subnuclei, often with terminal dendrites ending in the ventral spinocerebellar tract. The distal sections of the primary and higher-order dendrites exhibited rich varicosities, sometimes with dendritic spines. Axons of some labeled neurons were traced all the way to the ventrolateral pons (vl-pons). These findings extend and generalize the classical definition of the pneumotaxic center to include extensive somatic-axonal-dendritic integration of complex descending and ascending respiratory information as well as nociceptive and possibly musculoskeletal and trigeminal information in multiple dl-pons and vl-pons structures in the rat.

Barosensitive neurons in the rat tractus solitarius and paratrigeminal nucleus: a new model for medullary, cardiovascular reflex regulation.

The nucleus of the solitary tract (NTS), a termination site for primary afferent fibers from baroreceptors and other peripheral cardiovascular receptors, contains blood pressure-sensitive neurons, some of which have rhythmic activity locked to the cardiac cycle, making them key components of the central pathway for cardiovascular regulation. The paratrigeminal nucleus (Pa5), a small collection of medullary neurons in the dorsal lateral spinal trigeminal tract, like the NTS, receives primary somatosensory inputs of glossopharyngeal, vagal, and other nerves. Recent studies show that the Pa5 has efferent connections to the rostroventrolateral reticular nucleus (RVL), NTS, and ambiguous nucleus, suggesting that its structure may play a role in the baroreceptor reflex modulation. In the present study, simultaneous recording from multiple single neurons in freely behaving rats challenged with i.v. phenylephrine administration, showed that 83% of NTS units and 72% of Pa5 units were baroreceptor sensitive. Whereas most of the baroreceptor-sensitive NTS and Pa5 neurons (86 and 61%, respectively) increased firing rate during the ascending phase of the pressor response, about 16% of Pa5 and NTS baroreceptor-sensitive neurons had a decreased firing rate. On one hand, the decrease in firing rate occurred during the ascending phase of the pressor response, indicating sensitivity to rapid changes in arterial pressure. On the other hand, the increases in neuron activity in the Pa5 or NTS occurred during the entire pressor response to phenylephrine. Cross-correlational analysis showed that 71% of Pa5 and 93% of NTS baroreceptor-activated neurons possessed phasic discharge patterns locked to the cardiac cycle. These findings suggest that the Pa5, like the NTS, acts as a terminal for primary afferents in the medullary-baroreflex or cardiorespiratory-reflex pathways.

Baroreceptor-sensitive neurons in the rat paratrigeminal nucleus.

The paratrigeminal nucleus (Pa5) is a small collection of medullary neurons localized in the dorsal lateral spinal trigeminal tract. Electrophysiological and anatomical studies showed functional Pa5 efferent connections to the rostroventrolateral reticular nucleus (RVL) and the nucleus of the solitary tract (NTS), both well-studied components of the baroreflex arch. Similarly to the NTS, the main site for termination of cardiovascular peripheral afferents, the Pa5 receives primary sensory inputs of glossopharyngeal and vagus nerves, which suggests that the Pa5 may play a role in the baroreceptor reflex modulation. Simultaneous recording from multiple single neurons in 10 freely behaving rats showed that 37% of recorded Pa5 neurons altered firing rates (35% increased and 2% decreased) during the peak arterial blood pressure response to i.v. phenylephrine. Forty two percent of the 84 identified Pa5 baroreceptor-excited neurons showed high correlation to cardiac cycle denoting the synchronous phasicity to fast changes of blood pressure. Autocorrelation analysis revealed that 48 pressure-sensitive and 55 nonpressure-sensitive neurons have periodical activities which were not directly linked to cardiac cycle. We suggest that the Pa5, a yet unknown component of the baroreflex pathway, may relay baroreceptor information to the NTS and by passing other components of the baroreceptor reflex arch, directly to sympathetic premotor neurons in the RVL.

Cardiovascular responses to sciatic nerve stimulation are blocked by paratrigeminal nucleus lesion.

The paratrigeminal nucleus (Pa5) receives primary sensory inputs from the vagus, glossopharyngeal, and trigeminal nerves and has efferent projections to the nucleus of the solitary tract (NTS), rostroventrolateral reticular nucleus (RVL), as well as to the nucleus ambiguus (Amb), lateral reticular (LRt), parabrachial (PB) and ventral posteromedial thalamic (VPM) nuclei, suggesting that it may play a significant role in cardiovascular responses to nociceptive stimuli. The aim of the present study was to evaluate the effects of unilateral lesions of the Pa5 on cardiovascular alterations induced by afferent somatic sensory nerve stimulation (SNS), also known as the somatosympathetic reflex (SSR). Cardiovascular responses were recorded in rats following either sham operation or unilateral lesions of the Pa5 with ibotenic acid. Mean arterial blood pressure (MAP) increased after SNS, which in sham-lesioned animals raised from 95 +/- 4 to 115 +/- 2 mmHg. Ipsilateral Pa5 lesion did not significantly reduce the pressor response to SNS (from 91 +/- 7 to 107 +/- 4 mmHg increase of baseline MAP). On the other hand, contralateral Pa5 lesion significantly reduced the response to SNS (from 99 +/- 5, to 104 +/- 2 mmHg). Sciatic nerve stimulation did not alter heart rate (HR) neither did ipsi- or contralateral Pa5 lesion HR baseline response level. These findings support a crucial role for the Pa5 in cardiovascular regulation, by relaying SSR input evoked by peripheral nerve stimulation.

Systemic site of action for pressor effect of angiotensin II injected into the fourth cerebral ventricle of rats?

Angiotensin II (ANG II) causes a systemic pressor effect when injected into the cerebral ventricles. In the rat fourth ventricle, the effective doses for the ANG II pressor effect are over 100 times larger than in the systemic circulation. Considering the discrepancy of doses, the possibility that ANG II may reach the systemic circulation and promote pressor effects, following injection into the fourth ventricle, was investigated. The effects on blood pressure of different vasoactive peptides that produce pressor responses when injected into the central nervous system were compared. Dose-response curves were obtained for intravenous or fourth cerebroventricular injections of ANG II, lysyl-vasopressin (LVP), bradykinin (BK), or endothelin-1 (ET-1). The ED50 ratios for intracerebroventricular/intraveneous injections were 110 for ANG II, 109 for LVP, 0.01 for BK, and approximately 0.4 for ET-1. In cross-circulation preparations, pressor responses occurred in the donor rat following injection into the fourth cerebral ventricle of the recipient animal, showing that effective doses of ANG II, administered to the fourth cerebral, reach the systemic circulation. The same results were obtained for the microinjection of 4 nmol of LVP into the fourth cerebral ventricle of recipient animals. High-performance reverse-phase liquid chromatography analyses of arterial blood showed that approximately 1% of the [125I]ANG II injected into the fourth cerebral ventricle may be recovered from the systemic circulation a few seconds after the microinjection. The systemic administration of the ANG II receptor antagonist losartan blocked the response to ANG II injected into the fourth ventricle whereas antagonist administration in the same ventricle did not. Angiotensin injections into the lateral ventricle produced pressor responses that were reduced by antagonist administration to the same ventricle but not by systemic administration of the antagonist. The data suggest that the pressor effect resulting from ANG II or LVP injections into the fourth cerebral ventricle may be due to the action of this peptide in the systemic circulation. On the other hand, the pressor effect due to ANG II microinjection into the lateral ventricle apparently results from the direct stimulation of central periventricular structures.