Development of dendritic tonic GABAergic inhibition regulates excitability and plasticity in CA1 pyramidal neurons.

Synaptic plasticity rules change during development: while hippocampal synapses can be potentiated by a single action potential pairing protocol in young neurons, mature neurons require burst firing to induce synaptic potentiation. An essential component for spike timing-dependent plasticity is the backpropagating action potential (BAP). BAP along the dendrites can be modulated by morphology and ion channel composition, both of which change during late postnatal development. However, it is unclear whether these dendritic changes can explain the developmental changes in synaptic plasticity induction rules. Here, we show that tonic GABAergic inhibition regulates dendritic action potential backpropagation in adolescent, but not preadolescent, CA1 pyramidal neurons. These developmental changes in tonic inhibition also altered the induction threshold for spike timing-dependent plasticity in adolescent neurons. This GABAergic regulatory effect on backpropagation is restricted to distal regions of apical dendrites (>200 µm) and mediated by α5-containing GABA(A) receptors. Direct dendritic recordings demonstrate α5-mediated tonic GABA(A) currents in adolescent neurons which can modulate BAPs. These developmental modulations in dendritic excitability could not be explained by concurrent changes in dendritic morphology. To explain our data, model simulations propose a distally increasing or localized distal expression of dendritic α5 tonic inhibition in mature neurons. Overall, our results demonstrate that dendritic integration and plasticity in more mature dendrites are significantly altered by tonic α5 inhibition in a dendritic region-specific and developmentally regulated manner.

Role for the pineal and melatonin in glucose homeostasis: pinealectomy increases night-time glucose concentrations.

The effects of melatonin on glucose metabolism are far from understood. In rats, the biological clock generates a 24-h rhythm in plasma glucose concentrations, with declining concentrations in the dark period. We hypothesized that, in the rat, melatonin enhances the dark signal of the biological clock, decreasing glucose concentrations in the dark period. We measured 24-h rhythms of plasma concentrations of glucose and insulin in pinealectomized rats fed ad libitum and subjected to a scheduled feeding regimen with six meals equally distributed over the light/dark cycle and compared them with previous data of intact rats. Pinealectomy dampened the amplitude of the 24-h rhythm in plasma glucose concentrations in rats fed ad libitum, and abolished it completely in rats subjected to the scheduled feeding regimen, while plasma insulin concentrations did not change under both conditions. Pinealectomy abolished the nocturnal decline in plasma glucose concentrations irrespective of whether rats were fed ad libitum or subjected to the scheduled feeding regimen. Melatonin replacement restored 24-h mean plasma glucose concentrations in pinealectomized rats that were subjected to the scheduled feeding regimen but, interestingly, it did not restore the 24-h rhythm. Melatonin treatment also resulted in higher meal-induced insulin responses, probably mediated via an increased sensitivity of the beta-cells. Taken together, our data demonstrate that the pineal hormone, melatonin, influences both glucose metabolism and insulin secretion from the pancreatic beta-cell. The present study also demonstrates that removal of the pineal gland cannot be compensated by mimicking plasma melatonin concentrations only.

A daily rhythm in glucose tolerance: a role for the suprachiasmatic nucleus.

The suprachiasmatic nucleus (SCN), the biological clock, is responsible for a 24-h rhythm in plasma glucose concentrations, with the highest concentrations toward the beginning of the activity period. To investigate whether the SCN is also responsible for daily fluctuations in glucose uptake and to examine how these fluctuations relate to the rhythm in plasma glucose concentrations, SCN-intact rats and SCN-lesioned rats were injected intravenously with a glucose bolus at different time points. We found an increase in glucose uptake toward the beginning of the activity period, followed by a gradual reduction in glucose uptake toward the end of the activity period. The daily variation in glucose tolerance seemed not to be caused by fluctuations in insulin responses of the pancreas but by a daily variation in insulin sensitivity. Lesioning the SCN resulted in the disappearance of the daily fluctuation in glucose uptake and insulin sensitivity. Interestingly, SCN-lesioned rats showed an enhancement in glucose tolerance that could not be explained by higher insulin responses or enhanced insulin sensitivity. Therefore, these findings suggest a role for the SCN in insulin-independent glucose uptake. The present results further show that the daily rhythm in glucose tolerance follows the same pattern as the daily rhythm in plasma glucose concentrations. We hypothesized that the biological clock prepares the individual for the upcoming activity period by two separate mechanisms: increasing plasma glucose concentrations and making tissue more tolerant to glucose.

The suprachiasmatic nucleus generates the diurnal changes in plasma leptin levels.

At present it is not clear which factors are responsible for the diurnal pattern of plasma leptin levels, although the timing of food intake and circulating hormones such as glucocorticoids and insulin have both been proposed as independent determinants. In this study we show that ablation of the biological clock by thermal lesions of the hypothalamic suprachiasmatic nucleus (SCN) completely eliminates the diurnal pattern of plasma leptin levels. By contrast, removal of the diurnal corticosterone signal by adrenalectomy and corticosterone replacement did not affect diurnal plasma leptin levels. More importantly, removal of the nocturnal feeding signal by submitting the animals to a regular feeding schedule of six meals per day did not abolish the diurnal plasma leptin levels. However, both SCN lesions and the regular feeding schedule did cause an increase in the 24-h mean plasma leptin levels. As neither rhythmic feeding, insulin, or corticosterone signals can completely explain the diurnal plasma leptin rhythm, we conclude that biological clock control of the sympathetic input to the adipocyte is essential for regulation of the daily rhythm in leptin release.

Functional connections between the suprachiasmatic nucleus and the thyroid gland as revealed by lesioning and viral tracing techniques in the rat.

Frequent blood sampling via intraatrial cannula revealed daily rhythms of TSH and thyroid hormones in both male and female Wistar rats. Thermic ablation of the biological clock, i.e. the suprachiasmatic nucleus (SCN), eliminated the diurnal peak in circulating TSH and thyroid hormones. In addition, SCN lesions produced a clear decrease of 24-h mean T4 concentrations. A more pronounced effect of SCN-lesions on thyroid hormones, as opposed to TSH, suggested routes of SCN control additional to the neuroendocrine hypothalamopituitary-thyroid axis. Retrograde, transneuronal virus tracing was used to identify the type and localization of neurons in the central nervous system that control the autonomic innervation of the thyroid gland. In the spinal cord and brain stem, both the sympathetic and parasympathetic motorneurons were labeled. By varying the postinoculation survival time, it was possible to follow the viral infection as it proceeded. Subsequently, the pseudorabies virus (PRV) infected neurons in several brain stem cell groups, the paraventricular nucleus of the hypothalamus (PVN) and the central nucleus of the amygdala (second order labeling). Among PRV-infected neurons in the PVN were TRH-containing cells. Third order neurons were found in several hypothalamic cell groups, among which was the SCN. Therefore, we propose that the SCN has a dual control mechanism for thyroid activity by affecting neuroendocrine control of TSH release on the one hand and the autonomic input to the thyroid gland on the other.

Polysynaptic neural pathways between the hypothalamus, including the suprachiasmatic nucleus, and the liver.

The suprachiasmatic nucleus of the hypothalamus is responsible for a 24-h rhythm in basal glucose levels in the rat. The neural pathways used by the suprachiasmatic nucleus to mediate this rhythm in plasma glucose have not yet been identified. In the present study we examined whether there are any connections between hypothalamic centers, including the suprachiasmatic nucleus, and the liver, which is the main site for glucose production and storage. Transneuronal virus tracing from the liver showed that after injection of pseudorabies virus, specific neuronal cell populations in the central nervous system were labeled retrogradely, suggesting that specific sites in the central nervous system may control liver metabolism. First-order neurons belonged to the sympathetic and parasympathetic system, while second-order and third-order neurons were present in both the brainstem and hypothalamus. The presence of third-order neurons in the suprachiasmatic nucleus suggests an involvement of the biological clock in the neural control of the liver.

A suprachiasmatic nucleus generated rhythm in basal glucose concentrations.

The daily rhythm in feeding activity in mammals, as driven by the biological clock, largely determines the daily fluctuations in basal concentrations of glucose and insulin. To investigate a possible direct impact of the suprachiasmatic nucleus (SCN) on these parameters, we subjected intact rats and SCN-lesioned rats to a fasting regimen of 36 h, or to a scheduled feeding regimen of six identical meals equally distributed over the light:dark-cycle. Plasma profiles of glucose and insulin in rats during the final 24 h of the 36 h of fasting, and in rats subjected to the scheduled feeding regimen were compared to profiles in rats fed ad libitum. In rats fed ad libitum, in fasted rats and in rats subjected to a scheduled feeding regimen basal glucose concentrations showed a pronounced 24-h rhythm that was not found in rats that had been SCN-lesioned. Basal insulin levels showed a 24-h rhythm in 50% of the rats fed ad libitum and in 50% of the rats subjected to a scheduled feeding regimen; neither rhythms were present in SCN-lesioned rats. However, none of the fasted rats showed a 24-h rhythm in basal insulin concentrations. These data provide clear evidence that the SCN directly controls basal glucose concentrations independent of its influence on feeding activity. At the same time, we found no consistent evidence for a strong impact of the SCN on basal insulin concentrations.

Anatomical and functional demonstration of a multisynaptic suprachiasmatic nucleus adrenal (cortex) pathway.

In view of mounting evidence that the suprachiasmatic nucleus (SCN) is directly involved in the setting of sensitivity of the adrenal cortex to ACTH, the present study investigated possible anatomical and functional connections between SCN and adrenal. Transneuronal virus tracing from the adrenal revealed first order labelling in neurons in the intermedio-lateral column of the spinal cord that were shown to receive an input from oxytocin fibres and subsequently second-order labelling in neurons of the autonomic division of the paraventricular nucleus. The latter neurons were shown to receive an input from vasopressin or vasoactive intestinal peptide (VIP) containing SCN efferents. The true character of this SCN input to second-order neurons was also demonstrated by the fact that third-order labelling was present within the SCN, vasopressin or VIP neurons. The functional presence of the SCN-adrenal connection was demonstrated by a light-induced fast decrease in plasma corticosterone that could not be attributed to a decrease in ACTH. Using intact and SCN-lesioned animals, the immediate decrease in plasma corticosterone was only observed in intact animals and only at the beginning of the dark period. This fast decrease of corticosterone was accompanied by constant basal levels of blood adrenaline and noradrenaline, and is proposed to be due to a direct inhibition of the neuronal output to the adrenal cortex by light-mediated activation of SCN neurons. As a consequence, it is proposed that the SCN utilizes neuronal pathways to spread its time of the day message, not only to the pineal, but also to other organs, including the adrenal, utilizing the autonomic nervous system.

Anatomical demonstration of the suprachiasmatic nucleus-pineal pathway.

A polysynaptic pathway is proposed to transmit light information from the retina through the suprachiasmatic nucleus of the hypothalamus (SCN) to the pineal. In the present study, the powerful transneuronal tracer, pseudorabies virus (PRV), was used to provide a detailed description of this pathway. PRV injected into the pineal subsequently labeled the superior cervical ganglion, the intermediolateral column of the upper thoracic cord, the autonomic division of the paraventricular nucleus of the hypothalamus (PVN), and the SCN. Neurons in the autonomic division of the PVN were the only PRV-labeled neurons in the hypothalamus shown to receive input from the SCN as demonstrated by the presence of vasoactive intestinal polypeptide axonal contacts. This observation concurred with the presence of ventrally placed neurons in the SCN that could only be observed a day after the appearance of PVN-labeled neurons. Nevertheless the majority of the neurons were found in the dorsomedial position of the SCN, associated with the vasopressin-containing population of SCN neurons. Confocal laser scanning microscopy showed double-labeled neurons containing PRV and vasopressin or PRV and vasoactive intestinal polypeptide. Specificity of tracing was also established by prior removal of the superior cervical ganglion, resulting in a complete absence of the tracer but in the pineal. Thus, the present study provides the anatomical basis for circadian control of melatonin secretion.

Restricted daytime feeding modifies suprachiasmatic nucleus vasopressin release in rats.

The authors have shown previously that vasopressin (VP) release from suprachiasmatic nucleus (SCN) efferents in rats is important for the timing of the circadian activity of the hypothalamo-pituitary-adrenal (HPA) axis, resulting in a circadian rise in corticosterone at dusk. When meals are supplied at a fixed time during the light period, however, this normal circadian activity of the HPA axis is strongly modified. Under such a restricted feeding regimen, a corticosterone peak appears just before the daily meal in addition to the circadian corticosterone peak at dusk. This feeding-associated rise in corticosterone is regarded as an SCN-independent circadian rhythm because it is sustained in SCN-lesioned animals. Despite these previous results, the authors investigated a putative involvement of SCN-derived VP in the control of the prefeeding corticosterone peak by measuring the intranuclear release of VP in the SCN and plasma corticosterone levels in rats in ad libitum feeding conditions as well as in animals that were obliged to feed during a 2-h period in the middle of the light period. Restricted daytime feeding caused clear changes in the daily release pattern of VP from SCN terminals. Both a delayed onset of the diurnal rise and a premature decline of the elevated daytime levels were observed, but the acrophase of the VP rhythm was not phase shifted. Concerning the circadian corticosterone peak, no phase shift of its acrophase was observed either. It is concluded that (1) restricted daytime feeding does affect SCN activity, (2) intranuclear release of VP within the SCN is an important mechanism to amplify and synchronize the circadian rhythms as dictated by the light/dark-entrained circadian pacemaker, and (3) VP release observed in animals on restricted feeding is completely compatible with the previously proposed inhibitory action of SCN-derived VP on the HPA axis.

Immunocytochemical evidence for a diurnal rhythm of neurons showing colocalization of VIP with GRP in the rat suprachiasmatic nucleus.

The suprachiasmatic nucleus (SCN), which functions as a biological clock, contains several neuropeptides such as vasoactive intestinal peptide (VIP), peptide histidine isoleucine (PHI), and gastrin-releasing peptide (GRP). Studies from several laboratories have provided evidence for the coexistence of VIP with PHI and GRP, but reliable data about the proportions of colocalization and a possible diurnal rhythmicity are lacking. In the present study, we therefore aimed at studying these aspects. To this end, rats were killed by perfusion fixation during the middle of the day (Zeitgeber time [ZT] 7) and during the second part of the night (ZT 19). Coronal Vibratome sections through the SCN were double-immunolabeled for the presence of VIP and PHI or for VIP and GRP. Analysis of the sections was done by semi-quantitative confocal laser scanning fluorescence microscopy. It turned out that, in keeping with previous literature data, VIP and PHI always coexist at the cellular level. This was seen in all possible ratios, both during the day and at night. Part of these VIP/PHI-containing neurons (21%) and part of the GRP-containing neurons (33%) showed colocalization during the middle of the day. During the second part of the night, these percentages increased significantly to 28% and 40%, respectively. This increase in percentages was due to a significant, nocturnal increase of the number of profiles showing colocalization, in contrast to the number of profiles exclusively immunoreactive for VIP or GRP.

Novel environment induced inhibition of corticosterone secretion: physiological evidence for a suprachiasmatic nucleus mediated neuronal hypothalamo-adrenal cortex pathway.

Basal plasma ACTH and corticosterone levels are controlled by the suprachiasmatic nucleus (SCN), the site of the circadian pacemaker, resulting in a daily peak in plasma corticosterone and ACTH. The present study was carried out to investigate the mechanisms employed by the biological clock to control these hormones. Novel environment induced changes in plasma ACTH and corticosterone in intact and SCN-lesioned animals were employed as experimental approach. Placing intact animals in a new environment results in different plasma corticosterone and ACTH responses depending on the clock time of the stimulus. (1) Novel environment (2 h after onset of darkness (ZT14)) results in a fast decrease followed by an increase in corticosterone. This changing pattern in corticosterone secretion was not accompanied by any change in plasma ACTH, suggesting a direct neuronal control of the adrenal cortex. (2) In contrast, novel environment at 2 h after light onset (ZT2) results in a rapid increase in plasma ACTH. Regression analysis of the relation ACTH-corticosterone before and after stress shows a changed pattern at ZT2, although at that time still no significant correlation between ACTH and corticosterone was detected. AT ZT14 this correlation was only present after stress. (3) SCN lesioning results in low basal ACTH at all circadian times combined with elevated corticosterone levels. Here, a new environment results in an immediate increase in corticosterone without inhibition; ACTH also increases rapidly, but attains lower levels than at ZT2 in intact animals. (4) The present results therefore demonstrate SCN modulating corticosterone secretion by affecting ACTH secretion and changing the sensitivity of the adrenal cortex by means of a neuronal input.

Evidence from confocal fluorescence microscopy for a dense, reciprocal innervation between AVP-, somatostatin-, VIP/PHI-, GRP-, and VIP/PHI/GRP-immunoreactive neurons in the rat suprachiasmatic nucleus.

The rat suprachiasmatic nucleus (SCN) consists of several classes of neurons which can be identified by their transmitter content. Knowledge of putative interaction between these different cell types is essential in order to understand the possibilities of information processing within the SCN. The aim of the present study was therefore to obtain more information about the mutual innervation between the main cell classes in the rat SCN, viz. those containing the neuropeptides arginine vasopressin (AVP), vasoactive intestinal peptide (VIP), peptide histidine isoleucine (PHI), gastrin-releasing peptide (GRP) and somatostatin respectively. For this purpose, vibratome sections were double-immunolabelled for seven different peptide combinations and subsequently analysed by high-resolution confocal laser scanning fluorescence microscopy. Attention was focused on axosomatic appositions, the occurrence and frequency of which were quantitatively estimated. Our analysis of double-immunolabelled sections demonstrated that some of the VIP- and some of the GRP-immunoreactive nerve cells and endings showed colocalization. Assuming, on the basis of literature data, that VIP and PHI are always colocalized at the cellular level, the five main cell classes in the SCN appeared to be interconnected, at least axosomatically, in the following reciprocal way: AVP <--> VIP/PHI, AVP <--> GRP, AVP <--> somatostatin, somatostatin <--> VIP/PHI, somatostatin <--> GRP, VIP/PHI <--> GRP, VIP/PHI/GRP <--> GRP, VIP/PHI/GRP <--> VIP/ PHI. In addition to this heterologous axosomatic innervation, these cell groups also showed substantial homologous innervation. Supported by electron microscope data from the literature showing the existence of axodendritic synapses for some of these peptide combinations, our findings strongly suggest that the rat SCN comprises a complex synaptic network with strong interactive capabilities, which is probably a requisite for its biological clock function.

A diurnal rhythm of stimulatory input to the hypothalamo-pituitary-adrenal system as revealed by timed intrahypothalamic administration of the vasopressin V1 antagonist.

The mammalian suprachiasmatic nuclei (SCN) contain an endogenous pacemaker that generates daily rhythms in behavior and secretion of hormones. We hypothesized that the SCN imposes its circadian rhythm on the rest of the brain via a rhythmic release of its transmitters in its target areas. Previously, we demonstrated a pronounced inhibitory effect of vasopressin (VP), released from SCN terminals in the dorsomedial hypothalamus, on the release of the adrenal hormone corticosterone. In the present study, microdialysis-mediated intracerebral administration of the VP V1-receptor antagonist was used to pursue the study of the mechanisms underlying the circadian control of basal corticosterone release. Using timed administrations of the VP antagonist divided equally over the day/night cycle, we were able to uncover the existence of an additional stimulatory input from the SCN to the hypothalamopituitary-adrenal (HPA) axis. Peak activity of this stimulatory SCN input takes place during the second half of the light period, after the daily peak of VP secretion, with a delay of approximately 4-6 hr. In all likelihood, the inhibitory and stimulatory circadian input via separate mechanisms affects corticosterone release. Together, these two opposing circadian control mechanisms of the HPA axis enable a precise timing of the circadian peak in corticosterone release.

Differences in colocalization between Fos and PHI, GRP, VIP and VP in neurons of the rat suprachiasmatic nucleus after a light stimulus during the phase delay versus the phase advance period of the night.

Two groups of four rats each received a 15-minute light stimulus during the first part of the night (ZT14) and the second part (ZT19), respectively. After 45-60 minutes, the animals were killed by perfusion fixation. Adjacent Vibratome sections through the suprachiasmatic nucleus (SCN) were double-immunostained for the presence of peptide histidine isoleucine (PHI), gastrin releasing peptide (GRP) or vasoactive intestinal peptide (VIP) with Fos by using fluorophore-conjugated secondary antibodies. A few sections were triple-immunostained for PHI, GRP or VIP with vasopressin (VP) and Fos. Sections were analyzed with a confocal laser scanning microscope. It turned out that the ZT19 light stimulus induced 4.2 times more nuclear profiles in the SCN immunoreactive for Fos than the light stimulus given at ZT14. The SCN of control animals did not show any Fos immunoreactivity. After the ZT14 light stimulus, approximately 33% of the Fos profiles showed colocalization with a perikaryal profile immunoreactive for PHI, GRP or VIP, whereas at ZT19, this percentage had doubled to approximately 65%. After the light stimulus at ZT14, the relatively low Fos induction was numerically and proportionally most prominent in the PHI-immunoreactive perikarya. As compared with ZT14, the increase of Fos after the ZT19 light stimulus was most pronounced in the GRP-immunoreactive perikarya (21x) followed by VIP (15x) and PHI (5x). This outcome suggests that at least three different cell groups characterized by, respectively, PHI alone, GRP, and VIP fully or partly colocalized with PHI, play a prominent role during light-induced phase shifts: the PHI neurons during light-induced phase delays, the GRP and VIP/(PHI) neurons during light-induced phase advances.

Colocalization of gamma-aminobutyric acid with vasopressin, vasoactive intestinal peptide, and somatostatin in the rat suprachiasmatic nucleus.

The seemingly contradictory observations in previous publications that gamma-aminobutyric acid (GABA) is detected in all cell bodies of the suprachiasmatic nucleus (SCN) and that terminals originating from the SCN are only 20-30% GABA positive prompted us to investigate whether this might be explained by a preference of colocalization in terminals of certain peptidergic neurons in the SCN or by a day/night rhythm in GABA synthesis. At three different circadian times, animals were perfusion fixed, and their SCNs were stained for vasopressin (VP), somatostatin (SOM), or vasoactive intestinal polypeptide (VIP). Subsequently, the number of GABA peptide-positive terminals was determined using GABA postembedding staining in ultrathin sections. It appeared that the highest percentage of colocalization with GABA was detected in VIP terminals (38%) and the lowest in VP terminals (15%). No differences in colocalization percentages could be observed in any parameter at any circadian time. In the dorsomedial hypothalamus, one of the target areas of the VP and VIP fibers from the SCN, a colocalization of GABA within VP and VIP terminals was found similar to that in the SCN. In the region of the somatostatin-containing neurons in the SCN, a number of axoaxonal contacts could be observed that sometimes exhibited synaptic specializations. In nearly all cases, the axoaxonic terminals contained GABA and/or SOM. The conclusion is that the high level of intrinsic GABAergic connections in the SCN represents a putatively powerful mechanism to synchronize or shut down the activity of the SCN. We discuss the possibility that, depending on the firing frequency of the neurons, the colocalization of GABA with all peptides under investigation allows for the selection of which transmitter is released, the peptidergic one or the amino acid.

Projections from and to the superior colliculus in the nucleus of the optic tract combined with postembedding GABA immunocytochemistry in the rabbit.

The reciprocal connection between the nucleus of the optic tract and the superior colliculus has been described for several species. In both nuclei GABA is an important transmitter and recently it has been shown that GABA plays a significant role in this connection. To investigate this reciprocal connection at the ultrastructural level in the rabbit, anterograde and retrograde tracers were injected in the superior colliculus. Ultrathin sections were incubated with anti-GABA to elucidate the presence of GABA in retrogradely labeled neurons or anterogradely labeled terminals in the nucleus of the optic tract. The present results show that GABA is present in several neurons projecting from the nucleus of the optic tract to the superior colliculus and that all neurons involved in this projections are contacted by GABA immunopositive terminals. In the projection from the superior colliculus to the nucleus of the optic tract GABA play a less significant role.

The spectral sensitivity of blue-sensitive pigeon cones: evidence for complete screening of the visual pigment by the oil-droplets.

The relative spectral sensitivity of blue-sensitive pigeon cones was determined using flicker-photometry based on mass ERG-responses from intact subjects during high-level light adaptation. The resulting spectral sensitivity curve can be mimicked by a P460 nomogram screened by an oil-droplet that absorbs 50% at 466 nm. Comparison of the curve with the nomogram indicates that on the average less than 10% of the light that excites the blue-sensitive cones is guided past or diffracted around their oil-droplets.

Photopic spectral sensitivities of the red and the yellow field of the pigeon retina.

The spectral sensitivities of the red field and the yellow field in the retina of the homing pigeon (Columba Livia) were determined on the basis of ERG responses. Between 450 and 550 nm the relative spectral sensitivity of the yellow field turned out to be higher than that of the red field. The results are in agreement with spectral sensitivity data, obtained by behavioural threshold procedures.