Cognitive modelling reveals distinct electrophysiological markers of decision confidence and error monitoring.

Is confidence in perceptual decisions generated by the same brain processes as decision itself, or does confidence require metacognitive processes following up on the decision? In a masked orientation task with varying stimulus-onset-asynchrony, we used EEG and cognitive modelling to trace the timing of the neural correlates of confidence. Confidence reported by human observers increased with stimulus-onset-asynchrony in correct and to a lesser degree in incorrect trials, a pattern incompatible with established models of confidence. Electrophysiological activity was associated with confidence in two different time periods, namely 350-500 â€‹ms after stimulus onset and 250-350 â€‹ms after the response. Cognitive modelling revealed that only the activity following on the stimulus exhibited the same statistical regularities as confidence, while the statistical pattern of the activity following the response was incompatible with confidence. It is argued that electrophysiological markers of confidence and error awareness are at least in parts distinct.

The folded X-pattern is not necessarily a statistical signature of decision confidence.

Recent studies have traced the neural correlates of confidence in perceptual choices using statistical signatures of confidence. The most widely used statistical signature is the folded X-pattern, which was derived from a standard model of confidence assuming an objective definition of confidence as the posterior probability of making the correct choice given the evidence. The folded X-pattern entails that confidence as the subjective probability of being correct equals the probability of 0.75 if the stimulus in neutral about the choice options, increases with discriminability of the stimulus in correct trials, and decreases with discriminability in incorrect trials. Here, we show that the standard model of confidence is a special case in which there is no reliable trial-by-trial evidence about discriminability itself. According to a more general model, if there is enough evidence about discriminability, objective confidence is characterised by different pattern: For both correct and incorrect choices, confidence increases with discriminability. In addition, we demonstrate the consequence if discriminability is varied in discrete steps within the standard model: confidence in choices about neutral stimuli is no longer .75. Overall, identifying neural correlates of confidence by presupposing the folded X-pattern as a statistical signature of confidence is not legitimate.

Should metacognition be measured by logistic regression?

Are logistic regression slopes suitable to quantify metacognitive sensitivity, i.e. the efficiency with which subjective reports differentiate between correct and incorrect task responses? We analytically show that logistic regression slopes are independent from rating criteria in one specific model of metacognition, which assumes (i) that rating decisions are based on sensory evidence generated independently of the sensory evidence used for primary task responses and (ii) that the distributions of evidence are logistic. Given a hierarchical model of metacognition, logistic regression slopes depend on rating criteria. According to all considered models, regression slopes depend on the primary task criterion. A reanalysis of previous data revealed that massive numbers of trials are required to distinguish between hierarchical and independent models with tolerable accuracy. It is argued that researchers who wish to use logistic regression as measure of metacognitive sensitivity need to control the primary task criterion and rating criteria.

Metacognitive sensitivity of subjective reports of decisional confidence and visual experience.

Previous studies provided contradicting results regarding metacognitive sensitivity estimated from subjective reports of confidence in comparison to subjective reports of visual experience. We investigated whether this effect of content of subjective reports is influenced by the statistical method to quantify metacognitive sensitivity. Comparing logistic regression and meta-d in a masked orientation task, a masked shape task, and a random-dot motion task, we observed metacognitive sensitivity of reports regarding decisional confidence was greater than of reports about visual experience irrespective of mathematical procedures. However, the relationship between subjective reports and the logistic transform of accuracy was often not linear, implying that logistic regression is not a consistent measure of metacognitive sensitivity. We argue that a science of consciousness would benefit from the assessment of both visual experience and decisional confidence, and recommend meta-d as measure of metacognitive sensitivity for future studies.

A comparison between a visual analogue scale and a four point scale as measures of conscious experience of motion.

Can participants make use of the large number of response alternatives of visual analogue scales (VAS) when reporting their subjective experience of motion? In a new paradigm, participants adjusted a comparison according to random dot kinematograms with the direction of motion varying between 0° and 360°. After each discrimination response, they reported how clearly they experienced the global motion either using a VAS or a discrete scale with four scale steps. We observed that both scales were internally consistent and were used gradually. The visual analogue scale was more efficient in predicting discrimination error but this effect was mediated by longer report times and was no longer observed when the VAS was discretized into four bins. These observations are consistent with the interpretation that VAS and discrete scales are associated with a comparable degree of metacognitive sensitivity, although the VAS provides a greater amount of information.

Top-down dimensional weight set determines the capture of visual attention: evidence from the PCN component.

Visual search for feature singletons is slowed when a task-irrelevant, but more salient distracter singleton is concurrently presented. While there is a consensus that this distracter interference effect can be influenced by internal system settings, it remains controversial at what stage of processing this influence starts to affect visual coding. Advocates of the "stimulus-driven" view maintain that the initial sweep of visual processing is entirely driven by physical stimulus attributes and that top-down settings can bias visual processing only after selection of the most salient item. By contrast, opponents argue that top-down expectancies can alter the initial selection priority, so that focal attention is "not automatically" shifted to the location exhibiting the highest feature contrast. To precisely trace the allocation of focal attention, we analyzed the Posterior-Contralateral-Negativity (PCN) in a task in which the likelihood (expectancy) with which a distracter occurred was systematically varied. Our results show that both high (vs. low) distracter expectancy and experiencing a distracter on the previous trial speed up the timing of the target-elicited PCN. Importantly, there was no distracter-elicited PCN, indicating that participants did not shift attention to the distracter before selecting the target. This pattern unambiguously demonstrates that preattentive vision is top-down modifiable.

Competitive guided search: meeting the challenge of benchmark RT distributions.

Historically, visual search models were mainly evaluated based on their account of mean reaction times (RTs) and accuracy data. More recently, Wolfe, Palmer, and Horowitz (2010) have demonstrated that the shape of the entire RT distributions imposes important constraints on visual search theories and can falsify even successful models such as guided search, raising a challenge to computational theories of search. Competitive guided search is a novel model that meets this important challenge. The model is an adaptation of guided search, featuring a series of item selection and identification iterations with guidance towards targets. The main novelty of the model is its termination rule: A quit unit, which aborts the search upon selection, competes with items for selection and is inhibited by the saliency map of the visual display. As the trial proceeds, the quit unit both increases in strength and suffers less saliency-based inhibition and hence the conditional probability of quitting the trial accelerates. The model is fitted to data the data from three classical search task that have been traditionally considered to be governed by qualitatively different mechanisms, including a spatial configuration, a conjunction, and a feature search (Wolfe et al., 2010). The model is mathematically tractable and it accounts for the properties of RT distributions and for error rates in all three search tasks, providing a unifying theoretical framework for visual search.

Visual search for feature singletons: multiple mechanisms produce sequence effects in visual search.

Selection of a feature singleton target in visual search tasks, e.g., a red target among green distractors, is very fast--as if the target "popped out" of the display. Interestingly, reaction times (RTs) sometimes even decrease with an increase in the number of distractors (while keeping the presentation area fixed), i.e., there is a negative RT/display density relationship. Furthermore, repeating--versus changing--target-defining properties across trials also speeds up RTs. The present study investigated how display density influences two similar but dissociable types of such intertrial effects, namely (a) priming of pop-out (PoP), observed when the target-defining dimension is fixed, e.g., color, and only the features of the target and distractors, e.g., red and green, vary across trials and (b) the dimension-repetition effect (DRE), observed when both the features and dimensions of the target vary, e.g., from red circle (color) to blue square (shape target) among blue circles. Experiment 1 examined PoP magnitude with sparse (three-item) versus dense (36-item) displays in conditions in which the distractors' color either (a) varied, i.e., red target, green distractors versus green target, red distractors, or (b) it was fixed (blue). Significant PoP was observed only for sparse distractors conditions. Experiment 2 investigated the DRE magnitude across display densities with distractors always being fixed: Significant DREs of comparable magnitude were observed with both sparse and dense displays. This dissociation between the PoP and DREs suggests, first, the existence of multiple mechanisms of intertrial effects and, second, that PoP is specific to low target-distractor signal-to-noise ratios when the target fails to pop out.

Electromagnetic tracking of motion in the proximity of computer generated graphical stimuli: a tutorial.

Electromagnetic motion-tracking systems have the advantage of capturing the tempo-spatial kinematics of movements independently of the visibility of the sensors. However, they are limited in that they cannot be used in the proximity of electromagnetic field sources, such as computer monitors. This prevents exploiting the tracking potential of the sensor system together with that of computer-generated visual stimulation. Here we present a solution for presenting computer-generated visual stimulation that does not distort the electromagnetic field required for precise motion tracking, by means of a back projection medium. In one experiment, we verify that cathode ray tube monitors, as well as thin-film-transistor monitors, distort electro-magnetic sensor signals even at a distance of 18 cm. Our back projection medium, by contrast, leads to no distortion of the motion-tracking signals even when the sensor is touching the medium. This novel solution permits combining the advantages of electromagnetic motion tracking with computer-generated visual stimulation.

Simultaneous modeling of choice, confidence, and response time in visual perception.

How can choice, confidence, and response times be modeled simultaneously? Here, we propose the new dynamical weighted evidence and visibility (dynWEV) model, an extension of the drift-diffusion model of decision-making, to account for choices, reaction times, and confidence simultaneously. The decision process in a binary perceptual task is described as a Wiener process accumulating sensory evidence about the choice options bounded by two constant thresholds. To account for confidence judgments, we assume a period of postdecisional accumulation of sensory evidence and parallel accumulation of information about the reliability of the present stimulus. We examined model fits in two experiments, a motion discrimination task with random dot kinematograms and a postmasked orientation discrimination task. A comparison between the dynWEV model, two-stage dynamical signal detection theory, and several versions of race models of decision-making showed that only dynWEV produced acceptable fits of choices, confidence, and reaction time. This finding suggests that confidence judgments depend not only on choice evidence but also on a parallel estimate of stimulus discriminability and postdecisional accumulation of evidence. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Measures of metacognitive efficiency across cognitive models of decision confidence.

Meta-d'/d' has become the quasi-gold standard to quantify metacognitive efficiency because meta-d'/d' was developed to control for discrimination performance, discrimination criteria, and confidence criteria even without the assumption of a specific generative model underlying confidence judgments. Using simulations, we demonstrate that meta-d'/d' is not free from assumptions about confidence models: Only when we simulated data using a generative model of confidence according to which the evidence underlying confidence judgments is sampled independently from the evidence utilized in the choice process from a truncated Gaussian distribution, meta-d'/d' was unaffected by discrimination performance, discrimination task criteria, and confidence criteria. According to five alternative generative models of confidence, there exist at least some combination of parameters where meta-d'/d' is affected by discrimination performance, discrimination criteria, and confidence criteria. A simulation using empirically fitted parameter sets showed that the magnitude of the correlation between meta-d'/d' and discrimination performance, discrimination task criteria, and confidence criteria depends heavily on the generative model and the specific parameter set and varies between negligibly small and very large. These simulations imply that a difference in meta-d'/d' between conditions does not necessarily reflect a difference in metacognitive efficiency but might as well be caused by a difference in discrimination performance, discrimination task criterion, or confidence criteria. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Awareness in contextual cueing of visual search as measured with concurrent access- and phenomenal-consciousness tasks.

In visual search, context information can serve as a cue to guide attention to the target location. When observers repeatedly encounter displays with identical target-distractor arrangements, reaction times (RTs) are faster for repeated relative to nonrepeated displays, the latter containing novel configurations. This effect has been termed "contextual cueing." The present study asked whether information about the target location in repeated displays is "explicit" (or "conscious") in nature. To examine this issue, observers performed a test session (after an initial training phase in which RTs to repeated and nonrepeated displays were measured) in which the search stimuli were presented briefly and terminated by visual masks; following this, observers had to make a target localization response (with accuracy as the dependent measure) and indicate their visual experience and confidence associated with the localization response. The data were examined at the level of individual displays, i.e., in terms of whether or not a repeated display actually produced contextual cueing. The results were that (a) contextual cueing was driven by only a very small number of about four actually learned configurations; (b) localization accuracy was increased for learned relative to nonrepeated displays; and (c) both consciousness measures were enhanced for learned compared to nonrepeated displays. It is concluded that contextual cueing is driven by only a few repeated displays and the ability to locate the target in these displays is associated with increased visual experience.

Perceptual basis of redundancy gains in visual pop-out search.

The redundant-signals effect (RSE) refers to a speed-up of RT when the response is triggered by two, rather than just one, response-relevant target elements. Although there is agreement that in the visual modality RSEs observed with dimensionally redundant signals originating from the same location are generated by coactive processing architectures, there has been a debate as to the exact stage(s)--preattentive versus postselective--of processing at which coactivation arises. To determine the origin(s) of redundancy gains in visual pop-out search, the present study combined mental chronometry with electrophysiological markers that reflect purely preattentive perceptual (posterior-contralateral negativity [PCN]), preattentive and postselective perceptual plus response selection-related (stimulus-locked lateralized readiness potential [LRP]), or purely response production-related processes (response-locked LRP). As expected, there was an RSE on target detection RTs, with evidence for coactivation. At the electrophysiological level, this pattern was mirrored by an RSE in PCN latencies, whereas stimulus-locked LRP latencies showed no RSE over and above the PCN effect. Also, there was no RSE on the response-locked LRPs. This pattern demonstrates a major contribution of preattentive perceptual processing stages to the RSE in visual pop-out search, consistent with parallel-coactive coding of target signals in multiple visual dimensions [Müller, H. J., Heller, D., & Ziegler, J. Visual search for singleton feature targets within and across feature dimensions.

Visually guided pointing movements are driven by the salience map.

Visual salience maps are assumed to mediate target selection decisions in a motor-unspecific manner; accordingly, modulations of salience influence yes/no target detection or left/right localization responses in manual key-press search tasks, as well as ocular or skeletal movements to the target. Although widely accepted, this core assumption is based on little psychophysical evidence. At least four modulations of salience are known to influence the speed of visual search for feature singletons: (i) feature contrast, (ii) cross-trial dimension sequence and (iii) semantic pre-cueing of the target dimension, and (iv) dimensional target redundancy. If salience guides also manual pointing movements, their initiation latencies (and durations) should be affected by the same four manipulations of salience. Four experiments, each examining one of these manipulations, revealed this to be the case. Thus, these effects are seen independently of the motor response required to signal the perceptual decision (e.g., directed manual pointing as well as simple yes/no detection responses). This supports the notion of a motor-unspecific salience map, which guides covert attention as well as overt eye and hand movements.

Modelling visibility judgments using models of decision confidence.

How can we explain the regularities in subjective reports of human observers about their subjective visual experience of a stimulus? The present study tests whether a recent model of confidence in perceptual decisions, the weighted evidence and visibility model, can be generalized from confidence to subjective visibility. In a postmasked orientation identification task, observers reported the subjective visibility of the stimulus after each single identification response. Cognitive modelling revealed that the weighted evidence and visibility model provided a superior fit to the data compared with the standard signal detection model, the signal detection model with unsystematic noise superimposed on ratings, the postdecisional accumulation model, the two-channel model, the response-congruent evidence model, the two-dimensional Bayesian model, and the constant noise and decay model. A comparison between subjective visibility and decisional confidence revealed that visibility relied more on the strength of sensory evidence about features of the stimulus irrelevant to the identification judgment and less on evidence for the identification judgment. It is argued that at least two types of evidence are required to account for subjective visibility, one related to the identification judgment, and one related to the strength of stimulation.

Salience from the decision perspective: You know where it is before you know it is there.

In visual search for feature contrast ("odd-one-out") singletons, identical manipulations of salience, whether by varying target-distractor similarity or dimensional redundancy of target definition, had smaller effects on reaction times (RTs) for binary localization decisions than for yes/no detection decisions. According to formal models of binary decisions, identical differences in drift rates would yield larger RT differences for slow than for fast decisions. From this principle and the present findings, it follows that decisions on the presence of feature contrast singletons are slower than decisions on their location. This is at variance with two classes of standard models of visual search and object recognition that assume a serial cascade of first detection, then localization and identification of a target object, but also inconsistent with models assuming that as soon as a target is detected all its properties, spatial as well as non-spatial (e.g., its category), are available immediately. As an alternative, we propose a model of detection and localization tasks based on random walk processes, which can account for the present findings.

Contextual cueing of pop-out visual search: when context guides the deployment of attention.

Visual context information can guide attention in demanding (i.e., inefficient) search tasks. When participants are repeatedly presented with identically arranged ('repeated') displays, reaction times are faster relative to newly composed ('non-repeated') displays. The present article examines whether this 'contextual cueing' effect operates also in simple (i.e., efficient) search tasks and if so, whether there it influences target, rather than response, selection. The results were that singleton-feature targets were detected faster when the search items were presented in repeated, rather than non-repeated, arrangements. Importantly, repeated, relative to novel, displays also led to an increase in signal detection accuracy. Thus, contextual cueing can expedite the selection of pop-out targets, most likely by enhancing feature contrast signals at the overall-salience computation stage.

Positional priming of pop-out: a relational-encoding account.

Three experiments examined whether positional priming of pop-out is modulated by target salience. In Experiment 1, the singleton target appeared among variable numbers of distractors (2, 3, 5, 7; blocked presentation). While facilitation of target locations was not influenced by distractor number, inhibition of distractor locations was evident only with two distractors in the display. In Experiment 2, 3- and 6-item displays, with 2 and 5 distractors, respectively, were intermixed rather than blocked. It was found that, when the majority (but not the minority) of trials contained 3-item displays, there was carryover of distractor location inhibition from 3- to 6-item displays (but not vice versa). In Experiment 3 (1 target, 2 distractors), inhibitory priming of distractor locations could even be evoked when the current target was presented at an empty but merely expected distractor location in the previous trial. These findings argue that target salience is not an adequate account of positional priming. Instead they suggest that the relational encoding of the regular (triangle) stimulus arrangement contributes to positional priming.

Attentional capture by salient color singleton distractors is modulated by top-down dimensional set.

Three experiments examined whether salient color singleton distractors automatically interfere with the detection singleton form targets in visual search (e.g., J. Theeuwes, 1992), or whether the degree of interference is top-down modulable. In Experiments 1 and 2, observers started with a pure block of trials, which contained either never a distractor or always a distractor (0% or 100% distractors)--varying the opportunity to learn distractor suppression. In the subsequent trial blocks, the proportion of distractors was systematically varied (within-subjects factor in Experiment 1, between-subjects factor in Experiment 2)--varying the incentive to use distractor suppression. In Experiment 3, observers started with 100% distractors in the first block and were presented with "rare" color or luminance distractors, in addition to "frequent" color distractors, in the second block. The results revealed distractor interference to vary as a function of both the initial experience with distractors and the incentive to suppress them: the interference was larger without relevant practice and with a lesser incentive to apply suppression (Experiments 1-3). This set of findings suggests that distractor interference is top-down modulable.

Dimension- and space-based intertrial effects in visual pop-out search: modulation by task demands for focal-attentional processing.

Two experiments compared reaction times (RTs) in visual search for singleton feature targets defined, variably across trials, in either the color or the orientation dimension. Experiment 1 required observers to simply discern target presence versus absence (simple-detection task); Experiment 2 required them to respond to a detection-irrelevant form attribute of the target (compound-search task). Experiment 1 revealed a marked dimensional intertrial effect of 34 ms for an target defined in a changed versus a repeated dimension, and an intertrial target distance effect, with an 4-ms increase in RTs (per unit of distance) as the separation of the current relative to the preceding target increased. Conversely, in Experiment 2, the dimension change effect was markedly reduced (11 ms), while the intertrial target distance effect was markedly increased (11 ms per unit of distance). The results suggest that dimension change/repetition effects are modulated by the amount of attentional focusing required by the task, with space-based attention altering the integration of dimension-specific feature contrast signals at the level of the overall-saliency map.