However, during the third, deepest, Panobinostat in vivo choice, caudate activity was still associated with the values of both current choice alternatives but no longer with the value of the previously rejected root branch (Wunderlich et al., 2012a). This is exactly the pattern expected in a forward tree search during

goal-directed (model-based) decision making, where values related to distinct options are prospectively represented. Notably, these model-based effects were not evident in another basal ganglia structure, the putamen, which only encoded model-free values for extensively trained options at the time of choice. By contrast, when subjects were required to choose between an overtrained pair and half the tree, a situation requiring access to both model-based and model-free values, the caudate represented the planned target value of the decision tree, while activity in the putamen pertained solely to the value of the overtrained pair. This dissociation corresponds exactly to the response patterns of a model-free controller that depends on cached values (putamen) and a model-based

controller that depends on values calculated on the fly (caudate). Thus, when goal-directed and habit-based options compete, the activity in caudate and putamen covaried with planned and cached values even under situations where the relevant actions were not chosen. The findings fit snugly with an animal literature check details both in terms of anatomical dissociations as well as findings that highlight both systems act synergistically and in parallel (Wassum et al., 2009). In stark contrast, activity in vmPFC encoded the winning outcome of the choice process (chosen value), irrespective of whether this choice was based on a model-based or model-free value. Thus, vmPFC can access both model-based and model-free values, consistent with parallel, and independent, operation of model-based and model-free valuation systems. Simon and Daw

designed a different, spatial, task in order to examine model-based inference (Simon and Daw, 2011). Here, subjects navigated a maze consisting of a set of rooms connected by one-way doors Levetiracetam in order to get to goals; however, the structure of the maze changed randomly at every step, with the doors changing their allowed directions according to a small, fixed, probability. The constant change in the structure of the maze invited subjects to use model-based planning, and indeed their behavior was better fit by a model-based rather than a model-free method. Having pinned the behavior down, the authors were then in a position to study the neural representations of value signals associated with the planning task as well as other model-based quantities, such as the number of choices at the current and the next step in the maze (Simon and Daw, 2011).

The fact that

a runner can make an immediate alteration U0126 price in their footstrike pattern does not necessarily mean that this change is permanent. Adopting an FFS pattern places greater demands on the calf musculature.29 Thus, while patients were able to convert to an FFS pattern and significantly reduce their loading rates, additional training and conditioning of the lower leg and foot would be needed to maintain this pattern for longer duration runs. In addition, the increased load on the calf musculature, especially if it occurs suddenly, could put the runner at risk for muscle soreness or foot and ankle injuries resulting from overuse. This further supports the need for proper training, strengthening, and conditioning for a proper transition to an FFS pattern. Results of this study suggest that a runner who contacts the ground with less compliance,

or a higher vertical stiffness during IL, will exhibit a faster rise in VGRF, increasing Duvelisib purchase the loading rate. We found strong correlations between the change in VILS and the changes in VALR and VILR. Both VALR and VILR, as well as VILS significantly decreased from shod running to instructed BF running (Table 2). This suggests that runners decreased their vertical stiffness in order to eliminate impact transients and reduce loading rates. The relationship between stiffness and loading rate is not always apparent in the literature due to the technique used to compute vertical stiffness. Most studies use a constant stiffness, which does not model the transient of the VGRF, thus ignoring the high stiffness during IL. For example, Shih et al.24 reported a significant increase in loading rates between an FFS and an RFS pattern that had an impact transient early in stance. However stiffness was similar between groups. Similarly, Divert et al.30 found no difference in vertical stiffness between (-)-p-Bromotetramisole Oxalate shod and BF runners despite reporting that only three of 12 BF runners demonstrated an impact peak. This is because, in both studies,

stiffness was assessed as an average value across the entire loading phase. The constant stiffness model misrepresents the actual vertical stiffness in cases where an impact transient exists. However, the method introduced by Hunter13 and employed in this study is an important tool that provides a more accurate computation of stiffness, particularly during initial loading. In the current study, we reported a significant reduction in VILS between the shod and BF conditions (mean (|shod| − |BF|) = 14.7 ± 9.8 kN/m, p < 0.00001). However, had the simple constant stiffness model been applied to all steps, ignoring the impact transient, the results would have actually indicated the opposite. We would have seen an increase in VILS from the shod to instructed BF condition (shod: 23.2 ± 3.9 kN/m, BF: 24.9 ± 4.1 kN/m; mean(|shod| − |BF|) = −1.7 ± 2.2 kN/m; p < 0.00001). This increase in VILS during the BF condition would greatly misrepresent what is actually occurring.

The initial phylogenetic tree of the HA1 domain nucleotide sequences of each A-subtype or B-lineage was constructed with the PhyML software package version 3.0 [4] using GTR + I + Γ4. For this analysis the general time-reversible model with the proportion of invariant sites and the gamma distribution of among-site rate variation with four categories was estimated from the empirical data, determined by ModelTest [5] as the evolutionary model.

GARLI v0.961 [6] was run on the best tree from PhyML for 2 million generations to optimise tree topology and branch lengths for each virus A-subtype or B-lineage. The virus gene sequence accession numbers and their originating laboratories used in this report are listed in Table S1. A combination of antigenic and genetic data is routinely used to identify emergent antigenic variants. Antigenic cartography [7] was used to visualise the HI data. As discussed previously, the behaviour SP600125 manufacturer of A(H3N2) viruses in HA and HI assays has changed in recent years and their antigenic analyses have become more complex [8]. In particular, guinea pig RBC are now preferred for antigenic characterisation of current A(H3N2) PS-341 price viruses in HA and HI assays. To control for the possible participation of the virus NA in the agglutination of RBC, HI assays can also be performed in the presence of oseltamivir [9]. Virus neutralisation (plaque

reduction and microneutralisation) assays were performed in addition to HI tests for a subset of A(H3N2) viruses and a small number of A(H1N1)pdm09 viruses. In addition

to antigenic studies using post-infection ferret antisera, human serum panels obtained pre- and post-vaccination with seasonal influenza vaccine formulations were used to assess current vaccine coverage against representative recently circulating viruses. Serum panels for adults, elderly and paediatric populations received from Australia, China, Japan, the UK and the USA were tested where available. Only a relatively small number of A(H1N1)pdm09 viruses (392) were subjected to HI analysis by the WHO CCs from September 2012 to February 2013. The majority of these viruses remained below antigenically closely related to the vaccine virus A/California/7/2009 based on assays with post-infection ferret antisera and only 3.3% of these viruses had reduced titres of 8-fold or greater compared to titres against the homologous virus (Table 1). A high resolution phylogenetic tree of the HA genes was constructed and included 379 A(H1N1)pdm09 isolates collected through GISRS since February 2012 as shown in Fig. S1. While the phylogenetic tree of the A(H1N1)pdm09 HA gene can be divided into eight major genetic groups, the majority of viruses analysed for the VCM belonged to group 6 with the signature amino acid (AA) substitutions D97N, S185T and S451N in HA1 (Fig. 2, Fig. S1). Fewer viruses belonged to group 7 (signature AA substitutions N97D and A197T in HA1) were still present but fewer in number than in the previous reporting period.

Signals that drive motor learning can arise in different modalities, such as through vision or proprioception, and have differential importance in driving learning. For example visual feedback of hand trajectories is not required for adaptation

to novel stable (DiZio and Lackner, 2000, Scheidt et al., 2005 and Tong et al., 2002) or unstable dynamics (Franklin et al., 2007a). This result may not be unexpected because congenitally blind individuals are able to walk Bortezomib purchase and use tools (two examples of adaptation to unstable dynamics), and can adapt to the perturbing effects of a Coriolis force field (DiZio and Lackner, 2000). This demonstrates that visual feedback is not critical for adaptation to dynamics. Interestingly, when subjects were presented with no visual information regarding the errors

perpendicular to the movement direction, they could straighten their movements (adapting to the dynamics) but were unable to modify their movement direction and, therefore, unable to reach the original targets (Scheidt et al., 2005). This suggests that visual information appears to be responsible for learning the direction of the movement and path planning. Indeed, subjects without proprioception are able to adapt to visuomotor rotations (Bernier et al., 2006), suggesting that Selleck SCR7 the visual signal is enough for the remapping of movement direction planning. However, subjects without proprioception are unable to learn the correct muscle activation patterns to adapt to their self-produced joint-interaction torques during reaching (Ghez et al., 1995 and Gordon et al., 1995). Visual feedback does provide useful information for dynamical control, in particular to select different internal models of objects (Gordon et al., 1993). However, whereas visual feedback may predominately affect the learning and remapping of path planning, it appears that proprioceptive feedback predominately drives the learning and generalization of dynamics. Models of trial-by-trial adaptation have been developed to relate errors experienced on one trial to the update of internal representation of the forces or joint torques that will be produced on the subsequent trial (Kawato

et al., 1987, Scheidt et al., 2001 and Thoroughman and Shadmehr, Tryptophan synthase 2000). However, this approach is limited in several respects. First, it has been shown not to function in unstable environments, where the control of the limb impedance is required (Burdet et al., 2006 and Osu et al., 2003). Second, within the optimal control framework, motor learning should not be viewed as a process that only acts to reduce error. Indeed, other factors such as energy consumption (Emken et al., 2007), risk (Nagengast et al., 2010), and reward play a role in the determination of the manner in which adaptation occurs and may explain why subjects change to curved movements under certain circumstances (Chib et al., 2006 and Uno et al., 1989). Optimal control can predict the trajectories learned after force field adaptation (Izawa et al.

To verify that neural signals from the retina to SVS neurons and the SCN core are functional, we measured induction of the immediate early gene c-fos in Sox14gfp/gfp and control mice before and after acute light exposure (aL) ( Figures S3A and S3B). During the dark period,

the SCN and the SVS do not express c-Fos, but the gene is strongly upregulated 1 hr after light exposure during the subjective dark phase ( Figure S3B). We could detect no major differences in c-Fos levels between Sox14gfp/gfp and control mice upon acute light exposure ( Figure S3B). We conclude that Sox14gfp/gfp mice have not lost the ability to transduce signals from the retina to their diencephalic targets. To assess whether Sox14gfp/gfp mice have a functional SCN capable of generating an endogenous circadian rhythm, we measured the period of three well-known Panobinostat circadian behaviors and physiological responses that are controlled by the SCN: motor activity, feeding episodes, and core body temperature. The intrinsic period of

the clock becomes apparent under constant dark (DD) conditions. As expected, wild-type mice show a free-running circadian period shorter than 24 hr for all three parameters recorded. Sox14gfp/gfp mice displayed circadian behaviors with a free-running period similar to control mice ( Figures selleck chemical 7A–7C) (ambulations: wild-type 23.8, Sox14gfp/gfp 23.6; feeding episodes: wild-type 23.7, Sox14gfp/gfp 23.4; core body temperature: wild-type 23.7, Sox14gfp/gfp 23.5; median). We then reintroduced the light variable with the normal 12 hr light and 12 hr dark (LD) cycle. Under these conditions, both control mice and Sox14gfp/gfp mice adjusted their circadian rhythms in motor activity, feeding, and core body temperature,

giving periods very close to 24 hr ( Figures 7A–7C) (ambulations: wild-type Mephenoxalone 24.0, Sox14gfp/gfp 24.0; feeding episodes: wild-type 24.0, Sox14gfp/gfp 24.0; core body temperature: wild-type 23.8, Sox14gfp/gfp 24.0; median). The retained ability of Sox14gfp/gfp mice to respond to environmental light changes was also shown under a 6 hr LD phase advance experiment ( Figure S5B). Strikingly though, in Sox14gfp/gfp mice, the phase of all three circadian outputs did not align to the phase of the light cycle ( Figures 7E, 7G, 7I, and S5A–S5C). As a consequence of their advanced phase onset, mutant mice displayed increased ambulations during the L phase and decreased ambulations in the D phase compared to controls ( Figure 7D) (percentage, daily average ambulations in L phase: wild-type 14.9% ± 0.6%; Sox14gfp/gfp 47.1% ± 1.8%; average ± SEM). Similarly, both the onset of feeding and elevation of body temperature were phase advanced. This is shown in Figure 7, which compares feeding ( Figure 7F) and body temperature ( Figure 7H) in the 2 hr before D onset and the 2 hr before L onset for each genotype (percentage of daily average feeding: wild-type L 4.6% ± 0.6%, D 10.0% ± 1.

, 1996). Since then, much effort has been devoted to determine the presynaptic role of KARs and it is now widely accepted that functional presynaptic KARs play a crucial role in the control of neurotransmitter release

(Lerma, 2003). Indeed, it is now known that presynaptic KARs modulate neurotransmitter release in a bidirectional manner, not only at excitatory but also at inhibitory synapses. KAR activation modulates GABAergic transmission in a complex cellular and subcellular buy PD0332991 manner, and both depression and facilitation of GABA release have been reported (Figure 3). The question then arises as to which event takes preference over the other and under what circumstances? Early indications of KA-induced depression of inhibition in the hippocampus (Sloviter and Damiano, 1981) were confirmed by the demonstration that KARs can inhibit GABA release (Rodríguez-Moreno et al., 1997 and Vignes et al., 1998). The depression of inhibition induced was shown to be sensitive to PTx and to inhibitors of both PLC and PKC, leading to the postulate that KARs participated in unconventional events at presynaptic sites that most likely

involve a metabotropic signaling pathway rather than ion flux (Rodríguez-Moreno and Lerma, 1998). This idea was later PF-02341066 price supported by measuring GABA release in synaptosomes (Cunha et al., 1997, Cunha et al., 2000 and Perkinton and Sihra, 1999) and it has been observed in other structures such as the amygdala (e.g., Braga et al., 2004), neocortex (Ali et al., 2001), globus pallidus (Jin and Smith, 2007), and hypothalamic supraoptic nucleus (Bonfardin et al., 2010). However, CA1 interneurons become overactivated by exogenous KA through somatodendritic KARs, leading to the paradox of KA inducing Olopatadine both overflow (Frerking et al., 1998 and Cossart

et al., 1998) and inhibition of GABA release. Presynaptic and somatodendritic KARs seem to coexist, presenting distinct pharmacological profiles and subunit compositions and using different signaling pathways (Rodríguez-Moreno et al., 2000, Mulle et al., 2000, Christensen et al., 2004 and Maingret et al., 2005). Thus, while somatodendritic KARs mediate part of the synaptic input from Schafer collaterals, presynaptic KARs are activated by synaptically released glutamate and they reduce the inhibitory input to pyramidal cells (Min et al., 1999). Thus, KARs play a fundamental role in the performance of neuronal circuits, as exemplified in the hippocampus. As for other aspects of KAR activity, the mechanism by which KARs modulate inhibitory input to pyramidal neurons is not free of controversy.

Karila Helge Kasch Alice Kongsted Chia-Hua Kuo Brian C. Lau Kevin Laudner Kelly Laurson Lasse Lempainen Selleckchem Target Selective Inhibitor Library Heather C. Lench Jung-Charng Lin Shelly Linens Marc Lochbaum Inês Marques-Aleixo C. Mikael Mattsson Patrick McKeon Alison M. McManus William P. Meehan, III Goncalo Mendonca Christopher Mesagno Susumu Minamisawa Fiona J. Moola Christian Müller Tim Noakes Byeongsang Oh Fran Ortin A. Papaioannou S.M. Paranjape Brian Parr Oliver Pieske Danny Pincivero E.B.S. Ramanathan Sue Reeves Guus Reurink Joyce M. Richey Daniel Rodriguez Dieter Rosenbaum Leonard Rosenthall Hiroyuki Sasai Levy T. Shamah Ellen

Shanley Monique Simons Andrew J. Skalsky Raymond So Britt-Marie Stalnacke Nicholas Stanger Nicholas Stergiou Jason L. Talanian D. Thivel Keith Tolfrey Kumika Toma Philip Tomporowski Tom Tong E.G. Trapp Hans Tropp Evert Verhagen Arianne P. Verhagen Kirsten T. Verkooijen Judy Van Raalte H. Vernon Helen Walker Tristan Wallhead Chong-Wen Wang Henry Wang Yong Tai Wang Richard Weiler K.R. Westerterp Emma Wilmot Lei-Ting Xu Nobuo Yamaguchi Jin-Hong Yan Tongjian You Shuilian Yu Bohdanna

T. Zazulak Shi Zhou Full-size table Table options View in workspace Download as CSV “
“The cerebral cortex is the most recently evolved brain region in vertebrates and supports sophisticated sensory, motor, and cognitive functions in mammals. Despite its large size and functional diversification, the neocortex may have arisen from the duplication of stereotyped local circuits with subtle specializations in different cortical areas and Ergoloid species (Rakic, selleck products 2009). A major obstacle to understanding neural circuits in the cerebral cortex is the daunting diversity and heterogeneity of inhibitory interneurons (Markram et al., 2004). Compared with the more abundant glutamatergic projection neurons, GABAergic interneurons constitute only approximately 20% of cortical neurons, yet these interneurons are crucial in regulating the balance, flexibility, and functional architecture of cortical circuits (Klausberger and Somogyi, 2008 and Markram et al., 2004). GABAergic interneurons consist of a rich array

of cell types with distinct physiological properties, connectivity patterns, and gene expression profiles. Their diverse intrinsic, synaptic, and dynamic properties allow interneurons to generate a rich repertoire of inhibitory outputs (Jonas et al., 2004). Their distinct connectivity patterns ensure differential recruitment by appropriate inputs as well as strategic distribution of their outputs to stereotyped locations (e.g., specific cellular and subcellular targets) in cortical network (Buzsáki et al., 2004 and Somogyi et al., 1998). GABAergic interneurons also play key roles in various forms of network oscillations that provide spatial-temporal frameworks to dynamically organize functional neural ensembles (Bartos et al., 2007, Buzsáki, 2001 and Klausberger and Somogyi, 2008).

Koga et al.55 analyzed movement characteristics of 10 ACL injury cases in female team handball and basketball using the model-based manual image-matching technique. They estimated that injuries occurred about 40 ms after initial foot contact with the ground. Knee flexion and knee valgus increased during the first 40 ms after the initial foot contact with the ground, and that the knee was externally rotated at initial foot contact with find more the ground, and internally rotated during the first 40 ms after the initial foot contact. The investigators concluded that the valgus motion coupled with internal tibial rotation

under low knee flexion appeared to be important risk factors for ACL injury.55 However the measurement errors of the model-based manual image-matching technique were up to 11° in knee flexion angle, 13° in knee internal/external rotation angle, and

5° in knee varus/valgus angle.54 These significant measurement errors AZD6738 solubility dmso minimized the validity of this study. Another method to identify risk factors for ACL injury is to determine associations of injury risk factors with pre-injury movement characteristics through prospective cohort studies. In a prospective cohort study,56 205 adolescent soccer, basketball, and volleyball players were screened for lower extremity biomechanics in a drop landing task, and subsequently followed for 13 months. Nine ACL injuries (seven in soccer and two in basketball) occurred. Compared already to the non-injured players, the injured-players had increased knee abduction angles at initial contact, maximum knee abduction angles, maximum external knee abduction moments, peak vertical ground reaction

forces, maximum external hip flexion moments, and side-to-side knee abduction moment differences during landing, and decreased maximum knee flexion angles and stance time. Statistical analysis demonstrated that the knee abduction moment was the most sensitive factor to predict ACL injury with 75% specificity and 78% sensitivity. This was the first prospective cohort study in an attempt to screen jump-landing mechanics to identify biomechanical risk factors for ACL injury. However a small number of injuries, the late occurrence of the maximum knee valgus moment during the stance phase, a lack of horizontal deceleration in the testing task, and a lack of consideration of ACL loading mechanisms were identified as limitations of this study.23 Also a lack of cause-and-effect relationship between identified risk factors and the injury risk is another significant limitation of this type of prospective cohort study. Another study to prospectively identify risk factors for ACL injury was performed at three US military academies for 5 years.57 A total of 6124 cadets were screened for lower extremity biomechanics in a simulated stop-jump task. Ninety-eight cadets had ACL injuries after the screening.

, 1999 and Custer et al., 2006). Thus, decays of the basal synaptic transmission in EPAC−/− neurons could be also due to a reduction of Sv2b expression. Our electrophysiological recordings show that LTP is impaired in EPAC−/− neurons, indicating that functions

of EPAC proteins extend beyond the basal synaptic transmission. LTP is an activity-dependent long-lasting enhancement of synaptic transmission and is one major form of plasticity in the central neurons (Nicoll and Malenka, 1995). The induction (during tetanus) of LTP requires activation of post-synaptic NMDA receptors (Nicoll and Malenka, 1995) whereas expression of LTP or a late phase of LTP (i.e., 60 min after the induction), is involved of a series of the genomic responses for rapid new gene expression (Soderling and Derkach, 2000 and Kelleher et al., 2004). Consistent with this Adriamycin general idea, our data reveal that both NMDA

receptor channel activity at post-synaptic sites and LTP induction are normal in EPAC−/− neurons. Interestingly, our data demonstrate that a late phase of LTP is impaired in EPAC−/− mice and that this impairment correlates with a striking increase of miR-124 transcription, which caused Zif268 mRNA degradation. Zif268 is known as a critical transcriptional factor for stabilizing synaptic responses in LTP expression (Hall et al., 2000). Notably, either knockdown of miR-124 or expression of Zif268 is able to restore the capacity of EPAC−/− Bumetanide neurons to express LTP. Thus, EPAC signaling is directly linked with a key aspect of GS-7340 manufacturer the genomic responses including miR-124 transcription and Zif268 translation for LTP expression. Several rare nonsynonymous variants of EPAC genes were reported in patients with ASDs (Bacchelli et al., 2003) and one of them was found to be the loss of function mutation (Woolfrey et al., 2009). But the implications of these mutations in ASDs behaviors are not characterized yet. ASDs patients have been diagnosed as the abnormalities of social interactions and

mental retardation (Geschwind and Levitt, 2007, Kelleher and Bear, 2008 and Ramocki and Zoghbi, 2008). Additionally, the most human ASDs show poor language development and restricted and repetitive behaviors (Walsh et al., 2008 and Levy et al., 2009). In our present study, we show that genetic deletion of EPAC genes specifically defects the spatial learning and social interactions, suggesting a possibly mechanism by which the mutation of EPAC genes might reflect these ASDs behaviors. In conclusion, our results in the present studies provide genome-wide evidence that EPAC1 and EPAC2 proteins synergistically regulates miR-124 transcription and hence control Zif268 translation in the brain for processing spatial learning and social interactions.

The expression of this LTP involves presynaptic changes and requires AA signaling. Here, we demonstrate that excitatory synapses in the retina can undergo activity-dependent long-term synaptic plasticity. The absence of evidence for LTP in the retina had previously led to the idea that the lack Y-27632 clinical trial of long-term synaptic plasticity helps the stability of visual processing

in the retina. In recent years there are scattered studies showing that synapses in both adult and developing retinae are capable of undergoing long-lasting functional changes in response to intensive stimulation. In the adult goldfish retina, the transmission of reciprocal inhibitory synapses formed by amacrine cells

Crizotinib datasheet on BCs exhibits depolarization-induced enhancement for up to 10 min (Vigh et al., 2005). During the critical period of visual system development, the trafficking of AMPARs at mouse and rat BC-RGC synapses can be regulated by light illumination (Xia et al., 2006, 2007). In developing Xenopus tadpoles, long-term changes in synaptic AMPAR function at RGC dendrites can be induced by retrograde signaling from the optic tectum to retina ( Du and Poo, 2004; Du et al., 2009). Our present work directly demonstrates that during development, transmission of BC-RGC synapses in the zebrafish retina can be persistently potentiated by both repeated electrical and visual stimulations. This LTP is similar to the typical LTP found in central brain regions in both the time course and postsynaptic NMDAR dependency ( Lynch, 2004; Malenka and Bear,

2004). In the developing zebrafish retina, LTP can be induced at both ON and OFF inputs of ON-OFF, ON, and OFF RGCs. First, repetitive flash stimuli could induce LTP at BC-RGC synapses in all three subtypes of RGCs (six ON-OFF cells, one ON cell, and two OFF cells). Second, TBS could induce persistent enhancement of both ON (nine out of nine) and OFF (three out of eight) light responses among one ON and eight ON-OFF RGCs. Third, RFS could induce persistent enhancement of both ON (nine out of ten) and OFF (five out of eight) light responses in RGCs. Please note that these data suggest that ON synapses nearly on RGCs are more prone to undergo potentiation than OFF synapses. In mammals some subtypes of RGCs do not undergo dramatic developmental remodeling of their dendritic processes, but others do (Kim et al., 2010), implying that synaptic activity-induced LTP may only occur at some subtypes of RGCs. Transmitter release at the BC-RGC excitatory synapse, a typical ribbon synapse possessing high rates of exocytosis for transmitting graded potentials, is highly regulated (Sterling and Matthews, 2005; von Gersdorff et al., 1998; Wässle, 2004) by reciprocal inhibition from amacrine cells (Du and Yang, 2000; Vigh et al.