Performance-monitoring as a key function of cognitive control covers a wide range of diverse processes to enable goal directed behavior and to avoid maladjustments. Several event-related brain potentials (ERP) are associated with performance-monitoring, but their conceptual background differs. For example, the feedback-related negativity (FRN) is associated with unexpected performance feedback and might serve as a teaching signal for adaptational processes, whereas the error-related negativity (ERN) is associated with error commission and subsequent behavioral adaptation. The N2 is visible in the EEG when the participant successfully inhibits a response following a cue and thereby adapts to a given stop-signal. Here, we present an innovative paradigm to concurrently study these different performance-monitoring-related ERPs. In 24 participants a tactile time-estimation task interspersed with infrequent stop-signal trials reliably elicited all three ERPs. Sensory input and motor output were completely lateralized, in order to estimate any hemispheric processing preferences for the different aspects of performance monitoring associated with these ERPs. In accordance with the literature our data suggest augmented inhibitory capabilities in the right hemisphere given that stop-trial performance was significantly better with left- as compared to right-hand stop-signals. In line with this, the N2 scalp distribution was generally shifted to the right in addition to an ipsilateral shift in relation to the response hand. Other than that, task lateralization affected neither behavior related to error and feedback processing nor ERN or FRN. Comparing the ERP topographies using the Global Map Dissimilarity index, a large topographic overlap was found between all considered components.With an evenly distributed set of trials and a split-half reliability for all ERP components ≥.85 the task is well suited to efficiently study N2, ERN, and FRN concurrently which might prove useful for group comparisons, especially in clinical populations.
Article Figures and data Abstract Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract The Adolescent Brain Cognitive Development (ABCD) study is an unprecedented longitudinal neuroimaging sample that tracks the brain development of over 9–10 year olds through adolescence. At the core of this study are the three tasks that are completed repeatedly within the MRI scanner, one of which is the stop-signal task. In analyzing the available stopping experimental code and data, we identified a set of design issues that we believe significantly compromise its value. These issues include but are not limited to variable stimulus durations that violate basic assumptions of dominant stopping models, trials in which stimuli are incorrectly not presented, and faulty stop-signal delays. We present eight issues, show their effect on the existing ABCD data, suggest prospective solutions including task changes for future data collection and preliminary computational models, and suggest retrospective solutions for data users who wish to make the most of the existing data. Introduction The Adolescent Brain Cognitive Development (ABCD) is the largest and most comprehensive long-term study of brain development and child health in the United States (Casey et al., 2018). The study includes 11,878 youth and their families and aims to understand the environmental, social, genetic, and other biological factors that affect brain and cognitive development. This study was made possible by the Collaborative Research on Addiction at NIH (CRAN) including the National Institute of Drug Abuse, National Institute on Alcohol Abuse and Alcoholism, and National Cancer Institute in partnership with the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institute of Mental Health, National Institute of Minority Health and Health Disparities, National Institute of Neurological Disorders and Stroke, and the NIH Office of Behavioral and Social Sciences Research. CRAN granted $590 million in support to 21 research institutions across the United States to complete this study. At the core of the ABCD study are the structural and functional MRI brain scans that occur biennially for each participant. The initial baseline scans have all been completed. The study organizers chose to include three cognitive tasks to be presented during fMRI acquisition: the monetary incentive delay task (Knutson et al., 2000), the emotional N-back task (Barch et al., 2013), and the stop-signal task (Logan and Cowan, 1984). In this manuscript, we focus solely on the stop-signal task. We analyzed behavioral data from the baseline scan of 8,464 of the 11,878 participants. This subset resulted from the following exclusions. First, we attempted to download 8,811 participants from the ‘FastTrack Recommended Active Series’ from the NIMH Data Archive, which has gone through some quality assurance for the associated imaging files by the ABCD study organizers. Of these, 8,808 files were successfully downloaded. Of these, 8,734 files contained stop-signal task data that were encoded and formatted in text or csv format with header columns and each row representing one task trial. Finally, 270 subjects were removed who did not have two complete runs with 180 trials each, leaving us with a total of 8,464 complete datasets. We did not apply any other performance-based exclusion, including the ABCD study’s performance flag (Garavan et al., 2020), as this would have obscured our ability to evaluate potential issues (e.g., it would have masked Issue 3, see below). The stop-signal task is a primary paradigm used to understand response inhibition. It involves making a choice response to a go stimulus but attempting to stop that response when an infrequent stop signal occurs after a stop-signal delay (SSD). The dominant theoretical framework for understanding and interpreting stopping tasks is the independent race model (Logan et al., 2014; Logan and Cowan, 1984), which assumes that a go process begins when a go stimulus occurs and races independently against a stop process that begins when the stop stimulus occurs. The go process finishing first results in a stop failure (an overt response), whereas the stop process finishing first results in stop success (no response). The ABCD study is laudable for many reasons, not least of which is the dedication to making their data and materials openly available. This includes but is not limited to experimental code, trial-by-trial behavioral performance in each task, functional neuroimaging data, and structural neuroimaging data. Our research group aimed to use these data to understand the behavioral and neural underpinnings of response inhibition. In doing so, we first evaluated the experimental code and behavioral performance in the stop-signal task. During our analyses, we found a nexus of issues with the ABCD experimental code and behavioral data. The issues are as follows: different go stimulus durations across trials, the go stimulus is sometimes not presented, faulty SSD, different stop-signal durations for different SSDs, non-uniform conditional trial probabilities, trial accuracy incorrectly coded, SSD values start too short, and low stop trial probability. We judge these issues to vary from fundamental (e.g., different go stimulus durations across trials) to more minor (e.g., low stop trial probability). Indeed, some of the minor issues may reflect intentional design choices (e.g., low stop trial probability), but if so, we believe that those choices are suboptimal (for reasons that we lay out below). Additionally, we believe that the most fundamental issues are incontrovertible errors, and in the aggregate, we believe these eight issues significantly compromise the utility of the stopping data in this study for understanding the construct of response inhibition. We structure this paper as a series of issues. We order the issues roughly from what we judge to be most to least fundamental. For each, we outline the issue, demonstrate its effects in the ABCD data, suggest prospective solutions to the study organizers for future data collections, and suggest retrospective solutions to data users who want to make the most of the existing data. Also, we propose three preliminary computational frameworks that may be able to capture the violations that result from the first design issue. Since the original submission of this manuscript, a rebuttal preprint manuscript has been submitted from a group of authors involved in the design of the ABCD stop-signal task (Garavan et al., 2020). In short, we stand behind the conclusions of the present manuscript and found no evidence or argumentation within their rebuttal that meaningfully challenges any of our manuscript’s conclusions; importantly, their manuscript did not dispute the presence of any of the specific issues raised below. We would like to thank Garavan et al., 2020 for pointing out an error in our code, which we corrected on July 17, 2020 (https://www.biorxiv.org/content/10.1101/2020.05.08.084707v4). Before we lay out the first issue, we will first break down the trial structure of the ABCD stop-signal task (see Figure 1). 5/6 of all trials are go trials, in which a subject sees a go stimulus (a rightward- or leftward-pointing arrow) and makes one of two speeded responses based upon the direction of the arrow. The go stimulus is removed from the screen after 1000 ms or when a response occurs, whichever comes first. On 1/6 of all trials, this go stimulus is replaced with the stop signal (a vertical arrow) after the go stimulus has been on the screen for the duration of the SSD or when a response occurs; the stop signal is then presented for 300 ms (but see Issue 4). Therefore, on go trials, the go stimulus is on the screen for 1000 ms or the response time (RT), whichever comes first, whereas on stop trials, the go stimulus is on the screen for SSD or RT, whichever comes first, and is never concurrently on the screen with the stop stimulus. Figure 1 Download asset Open asset Stop-signal task trial structure. Results Issue 1: different go stimulus durations across trials This brings us to our first and most fundamental issue: the go stimulus is presented for a much longer duration on go trials than on stop trials. Mean go stimulus duration on go trials was 569 ms (standard deviation [SD] = 105 ms), and mean go stimulus duration on stop trials was 228 ms (SD = 118 ms), so on average, subjects had 341 ms longer on go trials to apprehend the go stimulus (see Figure 2 for full distributions). Figure 2 Download asset Open asset Proportion of go stimulus durations on go and stop trials. This is important because the current models for understanding stopping performance require that the go process is the same on go and stop trials. The main dependent variable in the stop-signal task is stop-signal reaction time (SSRT), which quantifies the latency of inhibition, but the estimation of SSRT requires application of a model because there is no overt response associated with a successful stop and thus stopping latency cannot be directly measured. The independent race model assumes context independence, which means that the go process and its finishing time are not affected by the presentation of the stop signal. It also assumes stochastic independence, which means that the finishing times of the go and the stop processes are independent on any given trial. In this manuscript, we focus on the assumption of context independence. Context independence is essential for calculating SSRT because context independence allows one to assume that the full distribution of responses on go trials can stand in for the (unobservable) full distribution of go processes on stop trials. Additionally, violations of context independence contaminate other major dependent variables in the stop-signal task, including the inhibition function and stop-failure RT (Bissett et al., 2021; Logan and Cowan, 1984). Paradoxically, some existing models have suggested that an array of interacting excitatory and inhibitory neurons give rise to behavior that appears independent, perhaps because the interaction is brief but very strong (Boucher et al., 2007). However, here we exclusively focus on behavioral evidence of independence or lack thereof. In both simple (e.g., Froeberg, 1907) and choice (e.g., Kaswan and Young, 1965) reaction times, there is evidence that shorter duration stimuli yield slower responses. This effect of stimulus duration on response speed occurs even for stimuli presented for hundreds of milliseconds (Kahneman and Norman, 1964; Kaswan and Young, 1965). This relates to Bloch’s law (Bloch, 1885), which states that intensity and duration can be traded off for shorter duration stimuli, though this trade-off only occurs over the first 100 ms. For example, reducing stimulus duration by half is equivalent to reducing intensity by half. This also relates to Pieron’s law, which states that RTs decrease with stimulus intensity (Pieron, 1914). Therefore, having shorter duration go stimuli on stop trials than go trials is akin to having a lower intensity go stimulus on stop than go trials, which slows RT. Taken together, a long history of work suggests that (all else being equal) shorter duration visual stimuli tend to yield slower responses. In the ABCD stopping experiment, go stimuli are presented for much shorter durations on stop trials than on go trials, so the work described in the preceding paragraph suggests that the go process will be faster on go trials than on stop trials. However, in order to extract SSRT, one must make the assumption that the go process is the same on go and stop trials (i.e., context independence). Therefore, we expect that context independence is violated in the ABCD dataset, which would contaminate major dependent variables in the stop-signal task including SSRT estimates. Additionally, because go stimuli on stop trials are presented for a duration equal to SSD, the degree to which violations of context independence occur are likely to differ across SSDs. When SSD is short (e.g., 50 ms), context independence should be more severely violated because the difference in go stimulus duration between stop (e.g., 50 ms) and go (up to 1000 ms) trials is so large. Evidence for Issue 1 in ABCD data The primary way to evaluate context independence is to compare reaction times on go trials to reaction times on stop-failure trials (but note that the prediction of the race model concerning faster stop-failure than go responses is conditioned on both context independence and stochastic independence, Colonius and Diederich, 2018). If the former is longer than the latter, then context independence is taken to hold (Bissett et al., 2021; Logan and Cowan, 1984; Verbruggen et al., 2019). On average across all subjects, stop-failure RT (M = 456 ms, SD = 109 ms) was shorter than overt responses on go trials (M = 543 ms, SD = 95 ms, 95% confidence interval of the difference [85.9, 88.8]). However, for 524 of the 8,464 subjects (6.2% of all subjects), mean stop-failure RT was longer than mean RT on overt responses in go trials, suggesting that a subset of subjects violated context independence. Though note that the comparison of stop-failure and go RT is a conservative measurement that will only show violations of context independence if they are severe (Bissett et al., 2021). Additionally, it will only reveal slowing of the go process on stop trials, even though context independence could be violated by the go process being faster on stop than go trials (e.g., see our guessing model below). In order to further evaluate whether the go process is impaired on stop trials as a result of the shorter go stimulus, we compared choice accuracy on stop-failure trials with choice accuracy on all overt (non-omission) go trials. Stop-failure trials had much lower accuracy (79%) than overt go trials (90%), 95% confidence interval of the difference (10.6%, 11.1%). Additionally, choice accuracy on stop-failure trials was increasingly impaired at shorter SSD (see Figure 3). In other datasets, choice accuracy on stop-failure trials, even at shorter SSDs, tends to be similar to overt go trials (Bissett et al., 2021), which suggests that the impaired go accuracy on stop-failure trials in the present study results from the shorter go stimulus durations on stop trials in this task. Therefore, this lower choice accuracy is consistent with the go process being fundamentally impaired on stop trials compared to go trials, particularly at short SSDs, violating the assumption of context independence. Additionally, these violations of context independence manifest in an unprecedented way (i.e., lower choice accuracy on stop failure trials), showing that Design Issue 1 drives a new form of violations that go beyond any previous evidence for violations in stop-signal tasks (e.g., Bissett et al., 2021). Figure 3 Download asset Open asset Choice response accuracy on stop-failure trials across SSD. Note: 95% confidence intervals are presented as gray confidence bands. Individual-subject datapoints are presented in blue with lower alpha. Potential mechanisms underlying Issue 1 New computational models need to be developed to capture the Design Issue 1 violations in the ABCD data. Issue 1 embeds violations in the essential fabric of their task design, and existing models do not provide guidance on how to excise these violations or limit their contamination of the ABCD data. To our knowledge, all existing models for stopping assume that the parameters that generate the distribution of go processes take the same value on go and stop trials (e.g., Boucher et al., 2007; Logan et al., 2014; Logan et al., 2015; Matzke et al., 2017). Therefore, there is no existing model that can capture the ABCD stopping data and no valid way to compute SSRT from these data because the go process is different on go and stop trials (see Figure 3). This is most clear on 0 ms SSD trials, in which the go stimulus is not presented at all (see Issue 2), so subjects must be guessing and therefore the go process must be fundamentally different than when it is stimulus driven. Additionally, existing simulations that suggest SSRT estimates are robust to violations of independence (Band et al., 2003) do not evaluate the type of contamination introduced by Issue 1, namely a different set of processes generating the go process on go and stop trials. Instead, they show that SSRT is relatively robust to correlations between go RT and SSRT or correlations between SSD and SSRT, which are unrelated to Design Issue 1. There is evidence that violations of context independence can occur even without the introduction of Design Issue 1 (Akerfelt et al., 2006; Bissett et al., 2021; Colonius et al., 2001; Gulberti et al., 2014; Ozyurt et al., 2003); however, we believe that the violations in the ABCD dataset go beyond this existing work and will require distinct computational solutions for the following reasons. First, these previous violations occur in spite of carefully equated exogenous stimulus parameters, whereas the ABCD design encourages violations by presenting different exogenous stimulus parameters between go and stop trials. Second, as we mentioned above, the violations in the ABCD dataset manifest in an unprecedented way (i.e., lower choice accuracy on stop-failure trials) that go beyond previous evidence for violations (e.g., Bissett et al., 2021). This suggests that the generating mechanisms underlying violations in ABCD are fundamentally different than in other stopping datasets. Theoretical explanations that aim to account for violations, such as failures to trigger the stop process (Matzke et al., 2017) or variability in the speed or potency of inhibition (Bissett et al., 2021), focus on modifications to the stop process across trials. In contrast, greatly reduced choice accuracy on stop-failure trials compared to go trials indicates that the go process is modified across go and stop trials in the ABCD dataset. Therefore, even if future computational modeling efforts account for violations in datasets without design Issue 1, we do not believe that they will generalize to the violations in the ABCD, given that the violations in ABCD are driven by the idiosyncrasies of Issue 1. In order to begin the process of developing a new theoretical framework for understanding the violations of context independence in the ABCD dataset, we propose three computational models that aim to capture the violations in the ABCD dataset. All three instantiate adjustments to the independent race model (Logan and Cowan, 1984). First, the slowed go processing model suggests that the drift rate for the go process, which measures the speed of information processing, is slowed at shorter SSDs. We justify this because weaker exogenous drive from a shorter stimulus should yield slower and less effective information processing of that stimulus. Second, the guessing model suggests that the go process on stop trials is a mixture of guesses and stimulus-driven responses, such that all guesses at 0 ms SSD stop trials gradually give way to all stimulus-driven responses (i.e., no guesses) as SSDs lengthen. We justify this because guesses are necessary at 0 ms and choice accuracy of stop-failure trials at shorter SSDs remains very low as if subjects are mostly randomly guessing. Third, the confusion model suggests that subjects may be confused on how to respond at short delays, impairing both the go and the stop processes, which we instantiate as slower drift rates for both the go and the stop processes at shorter SSDs. We justify this because short SSD trials may violate subjects’ basic expectation of a clear and perceptible go stimulus that precedes the stop signal on stop trials, resulting in confusion and weaker drive to both behavioral responses. All three of these models can naturally explain the higher choice error rate on stop-failure trials at short SSDs than long SSDs, as reducing drift rate reduces accuracy and guesses have chance accuracy. All three also roughly match the empirical inhibitory function from the ABCD data (see Figure 4), though only the third confusion model captures the non-monotonicity at the 0 ms SSD. For these reasons, they are promising, though preliminary, mechanistic models to explain the violations of context dependence in the ABCD data. Fully establishing these models will require substantial additional work, including parameter fitting and recovery, model recovery, and model comparison. Figure 4 Download asset Open asset Inhibition functions from the real ABCD data and four simulated models. With these caveats in mind, we simulated data to estimate the degree of SSRT contamination that would arise from applying the independent race model (Logan et al., 2014; Logan and Cowan, 1984) and assuming context independence if instead each of these three models are the correct generating model for the go process on stop trials. We fit the observed grand mean go RT and SSRT distributions to get plausible estimates of each value, then we adjusted the pertinent parameters (e.g., slow the drift rate of the go process at short SSDs in the slowed go processing model) according to the three models specified above (see Materials and methods for additional details). Finally, we computed an SSRT for simulated data generated from the independent race model and each of the three proposed models above while making the assumption of context independence (which we know to be violated in our three alternative models). This would overestimate mean SSRT by 75 ms for the slower drift rate model (slower drift rate SSRT = 207 ms and independent race SSRT = 282 ms), underestimate SSRT by 61 ms with the mixture guessing model (mixture guessing = 343 ms), and overestimate mean SSRT by 22 ms for the confusion model (confusion SSRT = 260 ms). Additionally, the degree of estimation errors varies with SSD (see Figure 5). Therefore, mean SSRT estimates would be contaminated if any of these three models capture the go process on stop trials, but the researcher instead assumed context independence, as the SSRT estimation procedure requires. These models instantiate differences in the go process between stop and go trials (i.e., context dependence), but by assuming context independence, these differences in the go process are incorrectly ascribed to the stop process, contaminating SSRT estimates. Figure 5 Download asset Open asset SSRT estimates from four generating models (slowed go processing, guessing, confusion, and independent race) across SSDs if we assume context independence. However, as Garavan et al., 2020 point out, ‘Examining individual differences is a central goal of the ABCD study. Crucially, for the utility of the ABCD SSRT measure to be degraded as a measure of individual differences, violations of context independence must result in more than a shift in mean SSRT. Rather, the rank order of participants’ SSRT values must be substantially altered’. To examine individual differences, we created 8,207 simulated subjects that shared features from the 8,207 real ABCD subjects. We simulated go RTs based on real ABCD subjects’ performance and implemented three ways to determine SSDs in the simulation (ABCD weighted, fixed SSDs, and simulated tracking SSDs). Given the observed violations of context independence, we do not have trustworthy estimates of individual-subject SSRTs. In order to assign an SSRT value to each simulated subject, we sampled randomly from an SSRT distribution with a mean that equaled the observed ABCD grand mean but assumed four different amounts of between-subject variability (ranging from SD = 0–85 ms). This range of between-subject variability was informed by evaluating the 20 simple stopping conditions from a recent large-scale stopping study which had a mean between-subject SD of SSRT = 43 ms with a range of 28 ms–85 ms (Bissett et al., 2021). For each simulated subject, we computed SSRT (assuming context independence) separately based upon data generated from our four generating models (slowed go processing, guessing, confusion, and independent race). We then computed rank correlations of SSRT estimates for these 8,207 simulated subjects between the independent race model and each of our three proposed models, for our three SSD determination methods (ABCD weighted, fixed, and tracking SSD) and four values of between-subject variance in assumed SSRT (85 ms, 25 ms, 5 ms, and 0 ms) (see Table 1 for Results and Materials and methods for additional simulation details). Table 1 Rank correlations of SSRTs from the three alternative generating models (rows) with the independent race model across the three SSD determination methods (ABCDw = ABCD weighted, fixed, and tracking) and four SSRT standard deviation (SD) scales (85 ms, 25 ms, 5 ms, and 0ms). SD = 85ms (mean:0.93)SD = 25ms (mean:0.78)SD = 5ms (mean:0.63)SD = 0ms (mean:0.54)ABCDwFixedTrackingABCDwFixedTrackingABCDwFixedTrackingABCDwFixedTrackingConfusion0.8480.9730.9950.730.890.9650.820.8930.5540.8470.9030.021Slowed Go Processing0.9320.9760.9490.6610.8960.6770.1780.8870.270.0930.8950.183Guessing0.8830.9070.9510.7130.7650.7540.8130.880.3420.8480.9130.176 Rank correlations quantify the degree that SSRT is degraded as a measure of individual differences if we assume the independent race model when a different generating model (like our three proposed models) more appropriately characterizes the underlying data generating mechanism in the ABCD stopping task. If we assume high between-subject variability in SSRT (SD = 85 ms), then the rank correlations are largely preserved across generating models and SSD approaches (r range = 0.85–0.99, mean = 0.93). However, as between-subject variability in SSRT reduces (SD = 25 ms), the rank correlations decrease (r range = 0.66–0.97, mean = 0.78). At very low between-subject variability, (SD = 5 ms) the rank correlations can become very low (r range = 0.18–0.89, mean = 0.63). At the limit, if we assume no between-subject variability in SSRT and therefore between-subject variance is driven entirely by differences in SSD, go RT, and the different generating models, rank correlations continue to reduce but remain well above 0 on average (r range = 0.02–0.91, mean = 0.54). This suggests that differences in SSD distribution or go RT across subjects can inflate estimates of SSRT rank correlations in our simulations, and therefore SSRT individual differences may be more contaminated than the above rank correlations suggest. Taken together, these results show that design Issue 1 can degrade individual differences, with the degree of misestimation increasing as between-subject variability in SSRT decreases. To reiterate, these are only preliminary model proposals and require more rigorous scrutiny. However, no existing models for stopping can capture the context dependence that is embedded in the fabric of the ABCD data by Design Issue 1. These preliminary simulations suggest that mean SSRT estimates are contaminated and individual differences may be contaminated by the Design Issue 1 violations of context independence, particularly if the true between-subject variability in SSRT is small. Prospective suggestions for Issue 1 In order to address Issue 1, we would recommend that the ABCD study organizers present the go stimulus for a fixed period of time on every trial, perhaps 1000 ms. When a stop signal occurs, it should not replace but it should be presented in addition to the go stimulus. If the study designers would like to keep all stimuli in the center of the screen, they could superimpose the stop stimulus around the arrow (e.g., a circle). However, as suggested by a reviewer, to avoid perceptual interactions the stop circle should be >1 degree of visual angle from the go arrow. Therefore, the go stimulus would be identical in form, size, and duration for all go and stop trials. This should eliminate any possibility that different go durations drive different go processes, violating context independence and contaminating dependent variables. Retrospective suggestions for Issue 1 The main reason that we have suggested that Issue 1 is the most fundamental is if one assumes that shorter go stimuli on stop trials yield slower, impaired go processes when compared with the longer go stimuli on go trials, which we believe is reasonable assumption given over 100 years of RT research (Bloch, 1885; Froeberg, 1907; Kahneman and Norman, 1964; Kaswan and Young, 1965; Pieron, 1914), then context independence is violated and major dependent variables in the stop-signal task are contaminated. Additionally, the violations of context independence, as measured by a decrease in choice accuracy on stop-failure trials, appear to go beyond existing evidence of violations in stopping data (Bissett et al., 2021). These violations may also influence common neuroimaging task contrasts like stop success versus go and stop failure versus go, as this introduces additional differences between trial types, including go stimulus duration and go stimulus reaction time, that will contaminate the ability of the contrast to isolate processes of interest like response inhibition. Given the above empirical evidence and simulations, and as suggested by our reviewers, unless the ABCD community shows that this design issue does not distort conclusions based upon SSRT estimates (or any other stop-signal measure), researchers should not use the ABCD dataset to estimate SSRTs and should use the neuroimaging data with caution. We also suggest two practi
Abstract Es wird gezeigt, wie unter Verwendung einer Xenon‐Hochdruck‐Lampe als Vergleichsstrahler Plasmatemperaturen bis 5000 K mit einem objektiven, kontinuierlich arbeitenden Verfahren gemessen werden können. Im Zusammenhang hiermit wird eine Methode entwickelt, regelbare Vergleichsstrahler mit Hilfe nur eines Temperatur‐Fixpunktes absolut zu eichen.
Cognitive control enables goal-oriented adaptation to a fast-changing environment and has a protracted development spanning into young adulthood. The neurocognitive processes underlying this development are poorly understood. In a cross-sectional sample of participants 8-19 years old (n = 108), we used blind source separation of EEG data recorded in a Flanker task to derive electrophysiological measures of attention and conflict processing, including a N2-like frontal negative component and a P3-like parietal positive component. Outside the recording session, we examined multiple behavioral measures of interference control derived from the Flanker, Stroop, and Anti-saccade tasks. We found a positive association between age and P3 amplitude, but no relationship between age and N2 amplitude. A stronger N2 was age-independently related to better performance on Stroop and Anti-saccade measures of interference control. A Gratton effect was found on the Flanker task, with slower reaction times on current congruent and better accuracy on current incongruent trials when preceded by incongruent as opposed to congruent trials. The Gratton effect on accuracy was positively associated with age. Together, the findings suggest a multifaceted developmental pattern of the neurocognitive processes involved in conflict processing across adolescence, with a more protracted development of the P3 compared to the N2.
Response inhibition refers to the suppression of prepared or initiated actions. Typically, the go/no-go task (GNGT) or the stop signal task (SST) are used interchangeably to capture individual differences in response inhibition. On the one hand, factor analytic and conjunction neuroimaging studies support the association of both tasks with a single inhibition construct. On the other hand, studies that directly compare the two tasks indicate distinct mechanisms, corresponding to action restraint and cancellation in the GNGT and SST, respectively. We addressed these contradictory findings with the aim to identify the core differences in the temporal dynamics of the functional networks that are recruited in both tasks. We extracted the time-courses of sensory, motor, attentional, and cognitive control networks by group independent component (G-ICA) analysis of electroencephalography (EEG) data from both tasks. Additionally, electromyography (EMG) from the responding effector muscles was recorded to detect the timing of response inhibition. The results indicated that inhibitory performance in the GNGT may be comparable to response selection mechanisms, reaching peripheral muscles at around 316 ms. In contrast, inhibitory performance in the SST is achieved via biasing of the sensorimotor system in preparation for stopping, followed by fast sensory, motor and frontal integration during outright stopping. Inhibition can be detected at the peripheral level at 140 ms after stop stimulus presentation. The GNGT and the SST therefore seem to recruit widely different neural dynamics, implying that the interchangeable use of superficially similar inhibition tasks in both basic and clinical research is unwarranted.
We all make mistakes—and when we do, it is a great opportunity for the brain to adjust what it is doing and to learn. To study how the brain detects and deals with errors, researchers have used caps equipped with sensors that can measure brain activity. One thing researchers have found using this method is that the brain creates a specific kind of brain activity when a person makes a mistake. This activity, called the error-related negativity or ERN, happens almost at the same time that the error is made. It is as if the brain already knows we are making a mistake within fractions of a second, before we are even aware of it. Where in the brain does this ERN come from? How does it help us learn? And how does it change as we develop from children to adults?
