During perceptual decisions, competing perceptual hypotheses are evaluated based on a combination of sensory data and prior knowledge. When observers are provided with predictive information about the correct choice, they exhibit behavioural biases favouring the more probable alternative, characterised by faster and more accurate decisions when the stimulus matches expectations (Summerfield and de Lange, 2014). How these expectations are encoded and integrated into the decision process is a critical question in cognitive science and a subject of ongoing debate (Deneve, 2012; Friston, 2010; Gold and Stocker, 2017; Heeger, 2017; Pouget et al., 2013; Press et al., 2020; Summerfield and de Lange, 2014; Teufel and Fletcher, 2020).

A key question dominating recent work in this field has been whether the influence of prior probability on our choice behaviour reflects purely strategic adjustments or also incorporates biases in the encoding or weighting of the sensory evidence that informs our decisions. Traditional accounts of the computational architecture of visual cortex emphasise a feedforward processing flow (Crick and Koch, 1998; Koch and Poggio, 1999) and tend to assume that early sensory processing exhibits a high degree of fidelity to the physical stimulus with little or no ‘cognitive penetration’ (Pylyshyn, 1999). Indeed, optimality research has demonstrated that prior knowledge can be incorporated into the decision process through strategic adjustments to decision criteria alone (Bogacz et al., 2006; van Ravenzwaaij et al., 2012) and mathematical models implementing such adjustments have been shown to comprehensively account for prior-informed behaviour (e.g. Leite and Ratcliff, 2011; Mulder et al., 2012; Mulder et al., 2014; Ratcliff and McKoon, 2008; Ratcliff and Smith, 2004). Neurophysiological research has provided further support for the role of criterion adjustments with numerous studies demonstrating elevated starting levels of motor preparation for the predicted response alternative (e.g. de Lange et al., 2013; Kelly et al., 2021). However, several modelling studies have indicated that these adjustments are accompanied by additional biases in the rate of evidence accumulation for the expected alternative at the decision-level (e.g. Dunovan et al., 2014; Hanks et al., 2011; Kelly et al., 2021; van Ravenzwaaij et al., 2012; Wyart et al., 2012). This raises the question of whether expectation-based effects in perceptual decision-making may in part arise from biases in the original encoding of the sensory evidence feeding the decision process.

An extensive recent literature spanning multiple species and neural assays has examined prior probability effects on sensory processing (de Lange et al., 2018; Feuerriegel et al., 2021b; Heilbron and Chait, 2018; Walsh et al., 2020) and a substantial number have provided compelling evidence of a variety of anticipatory and stimulus-evoked sensory-level modulations in humans (e.g. Aitken et al., 2020; Egner et al., 2010; Ekman et al., 2017; Esterman and Yantis, 2010; Grotheer and Kovács, 2015; Kok et al., 2012; Kok et al., 2013; Kok et al., 2014; Puri et al., 2009; Richter et al., 2018; Trapp et al., 2016). However, many of the key studies demonstrating expectation-based modulations of sensory activity in humans rely on voxel-based analyses of BOLD signals, where there is a limited understanding of the relationship between the macroscopic signal dynamics and the underlying processing of sensory evidence (Logothetis, 2008). There is evidence of expectation effects on sensory processing in electrophysiological research on non-human primates (e.g. Kaposvari et al., 2018; Meyer and Olson, 2011; Meyer et al., 2014; Ramachandran et al., 2017; Schlack and Albright, 2007; Schwiedrzik and Freiwald, 2017) and rodents (e.g. Findling et al., 2023; Fiser et al., 2016; Gavornik and Bear, 2014), but the evidence from human electrophysiology is more mixed (e.g. Aitken et al., 2020; den Ouden et al., 2023; Feuerriegel et al., 2018; Hall et al., 2018; Kok et al., 2017; Rungratsameetaweemana et al., 2018; Solomon et al., 2021; Stefanics et al., 2014; Tang et al., 2018). While some have reported expectation effects in humans using EEG/MEG, these studies either measured sensory signals whose relevance to the decision process is uncertain (e.g. Blom et al., 2020; Solomon et al., 2021; Tang et al., 2018) and/or used cues that were implicit or predicted a forthcoming stimulus but not the correct choice alternative (e.g. Aitken et al., 2020; Feuerriegel et al., 2021c; Kok et al., 2017). To assess whether prior probabilities modulate sensory-level signals directly related to participants’ perceptual decisions, we implemented a contrast discrimination task in which the cues explicitly predicted the correct choice and where sensory signals that selectively trace the evidence feeding the decision process could be measured during the process of deliberation.

The task was to discriminate the relative contrast of two overlaid gratings, so both gratings were equally task-relevant and the cues provided no spatial information. In addition, the task was designed to be difficult and to require extended deliberation, thus facilitating examination of sensory encoding responses before and during decision formation. We traced sensory encoding dynamics via the contrast-dependent steady-state visual-evoked potential (SSVEP; reviewed by Norcia et al., 2015), which is thought to originate in early visual cortex (Di Russo et al., 2007; Lauritzen et al., 2010; Vanegas et al., 2013) and is necessarily driven by units tuned to the stimulus, as only these units would be expected to respond at the specific frequency of the eliciting stimulus modulation. By flickering the two overlaid contrast gratings at different frequencies, it was possible to isolate a separate SSVEP for each grating and to measure their differential amplitude which would reflect the sensory quantity informing the perceptual choice. In addition to its excellent contrast sensitivity, the amplitude of SSVEP signals have been shown to predict choice behaviour on tasks where participants are asked to detect or discriminate stimulus contrast changes (e.g. Grogan et al., 2023; O’Connell et al., 2012; Steinemann et al., 2018), validating its characterisation as an index of evidence encoding.

We sought to determine if predictive cues lead to changes in sensory processing that could potentially contribute to the behavioural biases that expectations induce in perceptual decision-making. Given the more consistent observation of sensory-level prior probability modulation in studies of non-human animals that require extensive training, an additional goal of this study was to determine how any such sensory modulations might evolve as a function of extended task exposure. Here, we investigated if the magnitude of any modulation of sensory encoding associated with prior probabilities was enhanced as the participant became increasingly adept at contrast discrimination with practice and whether an emerging bias in sensory encoding may result in subtle adjustments to the incorporation of probabilistic information in the decision strategy.

We recorded 128-channel EEG data from 12 human participants performing a two-alternative forced-choice contrast discrimination task, where they decided which of two orthogonally oriented (±45° from vertical) overlaid gratings was presented with a higher contrast on discrete trials (see Figure 1). Across three to five sessions, participants completed 5750 trials on average. Each trial contained one of three predictive cues before evidence onset: valid cues correctly identified the tilt of the target stimulus on the subsequent trial; invalid cues indicated the opposite of the target tilt; and neutral cues provided no information about the likely answer. Contrast was individually titrated to achieve 70% accuracy and points were awarded/deducted for correct (+50 points) or error (-25 points) responses. Participants were instructed to maximise their score and respond as soon as they were quite sure of their decision by clicking the mouse button (left or right) corresponding to the tilt direction of the chosen grating. We evoked SSVEPs by separately 'frequency-tagging' the gratings by reversing their spatial phase at 20 and 25 Hz. This assignment of these frequencies to each grating was randomly counterbalanced across trials.

Figure 1

Download asset Open asset

The effect of the cue on behaviour

Participants exhibited the behavioural biases in reaction time and accuracy that are typically observed in response to prior probability cues. An initial full mixed effects model analysis of reaction time, including choice accuracy, revealed a significant interaction between Cue Validity and Choice Accuracy (F(2,66175)=92.23, p<0.001), which we followed up with separate mixed effect analyses for correct responses and error responses (Figure 2). The analysis of correct response reaction times revealed a main effect of the Cue Validity (F(2,50776)=64.41, p<0.001) and reaction times significantly increased with Task Exposure (F(1,50776)=17.92, p<0.001), but there was no interaction between the Cue Validity and Task Exposure (F(2,50776)=1.48, p=0.227). Compared to neutral cue trials, reaction times were significantly faster on valid cue trials (p<0.001; Δ=–0.043 ± 0.04) and significantly slower on invalid cue trials (p<0.001; Δ=0.085 ± 0.06).

Figure 2

Download asset Open asset

For error responses, there was also a main effect of the Cue Validity (F(2,15386)=36.47, p<0.001) and reaction times also significantly increased with Task Exposure (F(1,15386)=33.03, p<0.001), but again there was no interaction between the Cue Validity and Task Exposure (F(2,15385)=0.15, p=0.865). Compared to the neutral cue condition, error reaction times were significantly slower following a valid cue (p<0.001; Δ=0.084 ± 0.08) and significantly faster following an invalid cue (p<0.001; Δ=–0.047 ± 0.08).

A mixed effects analysis of choice accuracy indicated that there was a significant effect of the Cue Validity (F(2,66191)=692.46, p<0.001). However there was no significant effect of Task Exposure (F(1,66191)=0.23, p=0.63), and no interaction between the Cue Validity and Task Exposure (F(2,66191)=0.53, p=0.588). Compared to the neutral cue condition, participants’ accuracy significantly increased when they were given a valid cue (p<0.001; Δ=0.184 ± 0.01) and significantly decreased when they were given an invalid cue (p<0.001; Δ=–0.219 ± 0.007).

The trend of increasing reaction times and stable accuracy is likely at least partly attributable to the titration of task difficulty across sessions. While accuracy was maintained across testing sessions, a regression analysis showed that the marginal contrast was significantly reduced as a function of task exposure (F(1,58) = 7.17, p=0.01, R²=0.332). The mean marginal contrast across task exposure bins from first to last was 16.3% (SD = 6.4%), 13.08% (SD = 4.8%), 11.38% (SD = 4.1%), 11.5% (SD = 3.9%), 11.4% (SD = 3.7%). Despite the reduction in differential contrast across task exposure, task performance remained stable as participant scores did not change significantly across trial exposure bins (F(1,58) = 0.004, p=0.953).

The effect of the cue on the encoding of sensory evidence

Our SSVEP analyses centred on the ‘marginal SSVEP’: the difference between the amplitude of the signals generated by the target and non-target gratings. As part of an experimental manipulation whose effects will be examined in a separate paper, we introduced brief (150 ms) evidence pulses on a subset of trials in which the contrast difference was transiently increased or decreased (see Materials and Methods). Unless otherwise specified, the marginal SSVEP was measured in the window 680–975ms post evidence onset because that interval falls after the final pulse offset time (650 ms). Confirming that the SSVEP was a reliable neural index of the encoded contrast evidence, a Bayesian one-sample t-test indicated that the target (contrast-increase) SSVEP was significantly greater than the non-target (contrast-decrease) SSVEP (t(10) = 6.67, p<0.001, Figure 3A) in the neutral cue condition, with a Bayes Factor₁₀ of 500, which is considered very strong evidence.

Figure 3

Download asset Open asset

We first sought to characterise the influence of the marginal SSVEP signal on the outcome of the decision process. We used a binary logistic mixed effects model to predict accuracy on a trial-by-trial basis, using the amplitude of the marginal SSVEP for trials with reaction times greater than 680ms (Figure 3C). To control for its effects on choice accuracy the cue validity was also included in the model. There was a significant main effect of the marginal SSVEP (F(1,31186)=5.183, p=0.023) and Cue Validity (F(2,31186)=1289.273, p<0.001). The model coefficient indicated that accuracy significantly increased as the amplitude of the marginal SSVEP increased (β=0.009, p=0.023; see Figure 3C). A second mixed effects analysis revealed a significant effect of the marginal SSVEP (F(1,31186)=7.319, p=0.07) and Cue Validity (F(2,31176)=45.784, p<0.001) on reaction time. The model coefficient indicated that reaction times significantly decreased as the amplitude of the marginal SSVEP increased (β=–0.001, p=0.007; see Figure 3.D).

Several studies have indicated that expectations can evoke preparatory sensory activity before evidence onset (e.g. Kok et al., 2017; Trapp et al., 2016), which may itself constitute an early source of sensory evidence fed into the decision process (Feuerriegel et al., 2021a). To investigate this hypothesis, we measured the SSVEP amplitude in the pre-stimulus, baseline phase of the trial (–752:–214ms); comparing the marginal SSVEP representation of the cued stimulus (i.e. cued minus uncued SSVEP) to the marginal representation of the upcoming target stimulus (i.e. target minus non-target) in the neutral cue condition (Figure 4A). The cues did not produce a significant anticipatory modulation of sensory encoding (Cued vs Neutral, F(1,49289)=1.31, p=0.253).

Figure 4 with 1 supplement see all

Download asset Open asset

Addressing our primary hypothesis, a mixed effects analysis was conducted on the marginal SSVEP during evidence presentation to determine if the cue exerted any effect on the SSVEP representation of the sensory evidence (see Figure 4B, C). The analysis showed a main effect of the Cue Validity on the marginal SSVEP (F(2,44553)=4.21, p=0.015). The marginal SSVEP was significantly reduced following an invalid cue compared to either a valid cue (p=0.004; Δ=–0.148 ± 0.051) or a neutral cue (p=0.037; Δ=–0.136 ± 0.065). There was no significant difference between valid and neutral cue conditions (p=0.807; Δ=0.013 ± 0.051). A follow-up analysis on the amplitude of the target and non-target SSVEP was used to investigate whether the cue effect was primarily driven by a modulation of either or both of these signals. As expected, SSVEP amplitudes were significantly larger when elicited by Target compared to Non-Target gratings (F(1,89120)=4463.27, p<0.001) and the effect of the Cue Validity remained significant (F(2,89120)=3.48, p=0.031), but there was no Cue by Target/Non-Target interaction (F(2,89120)=2.92, p=0.054).

The Influence of task exposure on motor- and sensory-level effects

Several studies have shown that motor preparation signals are adjusted in response to predictive cues such that starting levels of motor preparation for the cued effector are elevated (e.g. de Lange et al., 2013; Kelly et al., 2021). We examined cue effects on motor preparation to compare the time course of their emergence to the patterns observed in the SSVEP. Visual inspection of the waveforms in Figure 5 suggests that there was strong lateralisation of motor signals during the baseline phase of each trial toward the expected alternative and this effect was evident in the earliest task exposure bin and remained largely stable across subsequent bins. This was confirmed by a mixed effects analysis of the baseline phase of the trial (–752:–215ms), which showed there was significantly greater MB lateralisation in favour of the expected alternative for cued compared to neutral trials (F(1,52540)=13.94, p<0.001), but no change in the extent of this lateralisation across Task Exposure (F(1,52540)=0.05, p=0.819) and no interaction between the Cue Condition and Task Exposure (F(1,52540)=0.23, p=0.631).

Figure 5

Download asset Open asset

Having established in the previous section that cues did modulate SSVEP amplitude, we conducted a second analysis to determine if this effect changed over the course of testing (Figure 5). A mixed effects analysis revealed a significant decline in marginal SSVEP as a function of Task Exposure (F(1,44550)=69.16, p<0.001). We attribute this decline to the per-session titration of task difficulty which resulted in a progressive reduction in the marginal contrast of the grating stimuli. We also observed a significant interaction between Cue Validity and Task Exposure driven by the fact that the cue validity effect on the marginal SSVEP only emerges in the later stages of testing (F(2,44550)=4.31, p=0.013). The model coefficients indicated that there was a significantly greater decline in amplitude as a function of task exposure in the invalid cue condition (β=–0.123, p=0.007). With the model accounting for this interaction, there was no longer the main effect of Cue Validity (F(2,44550)=1.20, p=0.302) that was observed in the previous analysis. Thus, whereas starting-levels of motor preparation were biased by the cues at the earliest stages of testing, the sensory-level modulations only emerged after substantial task exposure.

To test the possibility that the observed SSVEP modulation arises primarily after choice commitment (e.g. reflecting confirmation bias; see Figure 4—figure supplement 1), a mixed effects analysis was conducted on the difference between the pre- (–200:–50ms) and post-response marginal SSVEP (50:200ms). This ‘difference SSVEP’ increased with Task Exposure (F(2,42605)=4.03, p=0.045), but there was no main effect of Cue Validity (F(2,42605)=2.103, p=0.122). Furthermore, there was no interaction between Cue Validity and Task Exposure (F(2,42605)=2.38, p=0.093). To quantify the evidence that the cue effect was not driven by changes in the signal after the response, we ran Bayesian one-way repeated measures ANOVAs on the marginal SSVEP comparing the difference across cue conditions before and after the response. If the cue effect only emerged after the response, we would expect the difference between invalid and neutral or invalid and valid cues to increase in the post-response window. There was no compelling evidence of an increase in the effect when comparing invalid to neutral (BF₁₀=1.58) or valid cues (BF₁₀=0.32). As can be seen in the response-locked plot (Figure 4D), the amplitude of the grand average valid SSVEP is consistently greater than that of the invalid SSVEP before and after the response, indicating that the effect was not a purely post-choice phenomenon.

The effect of the cue on stimulus engagement

A priori, one might have expected that a valid cue should boost the marginal SSVEP compared to the neutral cue, but we observed no such effect, with amplitude instead significantly reduced following invalid cues relative to the other cue types. Given the high difficulty of the task (titrated to 70% accuracy for each session), we hypothesised that the presence/absence of cues may have impacted on the participants’ degree of engagement with the stimulus. That is, in the absence of any predictive information on neutral trials, participants must rely entirely on their ability to efficiently accumulate samples of evidence to respond correctly, and therefore they may deploy greater attentional resources in order to optimise sensory encoding. It has been well established that alpha-band oscillatory activity over occipito-parietal sites desynchronises during focussed attention and predicts performance on perceptual discrimination and detection tasks (Babiloni et al., 2006; Ergenoglu et al., 2004; Hanslmayr et al., 2007; Kelly and O’Connell, 2013; Kelly et al., 2006; Kelly et al., 2009; Linkenkaer-Hansen et al., 2004; O’Connell et al., 2009; van Dijk et al., 2008). This desynchronisation of alpha activity has been linked with activity in the dorsal attention network (Laufs et al., 2003; Laufs et al., 2006; Mantini et al., 2007; Sadaghiani et al., 2010; Scheeringa et al., 2009), and is thought to arise from the increased excitability in task-relevant processing regions and the suppression of task-irrelevant activity (Jensen and Mazaheri, 2010; Romei et al., 2008; Thut and Miniussi, 2009; Van Diepen et al., 2019). In a post-hoc analysis, we tested for greater task engagement in the neutral cue condition by measuring alpha-band oscillatory activity over a central occipito-parietal site in the pre-evidence window –410:–70ms (Figure 6). As with the MB analysis, the measurement window fell before evidence onset, allowing us to collapse the valid and invalid cue conditions into a single ‘cued’ condition. A mixed effect analysis showed a main effect of Cue Condition on alpha amplitude (F(1,52527)=10.10, P=0.001), such that there was significantly greater alpha desynchronisation in the neutral cue condition, consistent with increased task engagement in this condition.

Figure 6

Download asset Open asset

In the past decade, there has been a surge of interest in the role of expectation in perceptual processing, which has developed into a substantial literature reporting expectation-based modulations of neural activity across a variety of brain areas, species, paradigms, and recording techniques. This work has, in turn, spurred debate about the traditional characterisation of early sensory processing as insulated from cognitive influences. While some studies have presented evidence that probabilistic expectations induce behavioural biases without changing perceptual sensitivity (e.g. Bang and Rahnev, 2017; Rao et al., 2012; Rungratsameetaweemana et al., 2018), others have demonstrated anticipatory and/or stimulus-evoked sensory-level modulations (e.g. Aitken et al., 2020; Albright, 2012; Egner et al., 2010; Ekman et al., 2017; Esterman and Yantis, 2010; Kok et al., 2012; Kok et al., 2013; Kok et al., 2014; Kok et al., 2017; Puri et al., 2009; Rahnev et al., 2011; Richter et al., 2018; Trapp et al., 2016). However, this previous work either measured sensory signals whose relevance to the decision process is uncertain and/or used cues that predicted a forthcoming stimulus but not the correct choice alternative. To assess whether prior probabilities modulate sensory-level signals directly related to participants’ perceptual decisions, we implemented a task in which the cues explicitly predicted the correct choice and in which sensory signals that selectively trace the evidence feeding the decision process could be measured during the process of deliberation. Like previous studies (O’Connell et al., 2012; Steinemann et al., 2018), we found that the SSVEP closely tracked the differential contrast representing the sensory evidence in this task and was a significant predictor of choice accuracy and reaction time, as one would expect of an evidence encoding signal. We also found that the differential SSVEP responses were significantly modulated by predictive cues consistent with the proposal that prior probabilities are capable of modulating neural activity at early cortical stages of perceptual processing even when those expectations are specifically about the sensory features being decided upon. Crucially, however, this effect only emerged with substantial task exposure, suggesting that previous studies may have overlooked such sensory-level effects because they administered significantly fewer trials (e.g. den Ouden et al., 2023).

Several studies suggest that expectations may lead to the generation of anticipatory feature-specific sensory templates, which shape the processing of the subsequent stimulus (e.g. Blom et al., 2020; Ekman et al., 2017; Esterman and Yantis, 2010; Kok et al., 2014; Kok et al., 2017; Puri et al., 2009; Trapp et al., 2016). Here, we did not observe any significant anticipatory expectational modulations of sensory activity when measuring SSVEP amplitudes during a baseline interval between cue and evidence onset, during which the overlaid grating stimuli were shown at equal contrast. However, typically, studies that have detected preparatory activity did not present a baseline version of the expected stimulus in the window where the anticipatory modulation was recorded (e.g. Kok et al., 2014; Kok et al., 2017; Trapp et al., 2016) and the preparatory activity has been reported to be significantly weaker than the responses to physical stimuli. It is possible that this relatively subtle activity was obscured in the present study by the response to the presentation of physical grating stimuli during the pre-evidence baseline period. It is also not clear how far in advance of stimulus onset this kind of preparatory modulation can be expected to manifest. To our knowledge, the only estimate of this was offered by Kok et al., 2017 who found that the pre-stimulus activity only became predictive of the upcoming stimulus 40ms before stimulus onset. If that narrow window of predictive activity is representative of anticipatory modulations in general, the temporal resolution of the SSVEP would be insufficient to appropriately interrogate such an effect (as it relied on a 400ms window of activity). It is also plausible that our participants were sufficiently well-trained on the task timings that the influence of the prior was restricted to the period of evidence presentation without shaping the baseline phase where participants knew that the gratings were presented at equal contrast and had no reason to ‘expect’ otherwise. Indeed, some of the studies reporting anticipatory modulations used unpredictable stimulus onset times (e.g. Trapp et al., 2016) and there is some evidence that when participants are familiar with task timings, preparatory activity is initiated only when the predicted stimulus is expected to onset (e.g. Esterman and Yantis, 2010).

In contrast to the pre-evidence analysis, the cues did significantly impact on the marginal amplitude of the SSVEP following evidence onset, with significantly smaller responses following invalid cues compared to valid and neutral cue trials. This pattern is apparently at odds with a common observation in the expectation literature, that evoked responses to unexpected stimuli tend to be larger than responses to expected stimuli (Feuerriegel et al., 2021b; Walsh et al., 2020). These expectation suppression effects have been interpreted as consistent with predictive processing schemes, where unexpected sensory events provoke prediction error signals that call for a revision of perceptual hypotheses (Friston, 2005). These error signals are thought to emanate from the same superficial pyramidal cells that are primarily responsible for generating EEG signals (Cohen, 2017; Friston, 2009), further underlining the apparent discrepancy between these schemes and the present SSVEP results. An alternative account for expectation suppression effects, which is consistent with our SSVEP results, is that they arise, not from a suppression of expected activity, but from a ‘sharpening’ effect whereby the response of neurons that are tuned to the expected feature are enhanced while the responses of neurons tuned to unexpected features are suppressed (de Lange et al., 2018). On this account, the expectation suppression commonly reported in fMRI studies arises because voxels contain intermingled populations with diverse stimulus preferences and the populations tuned to the unexpected features outnumber those tuned to the expected feature. In contrast to these fMRI data, the SSVEP represents the activity of sensory units driven at the same frequency as the stimulus, and thus better isolates the feature-specific populations encoding the task-relevant sensory evidence. Therefore, according to the sharpening account, an invalid cue would have enhanced the SSVEP signal associated with the low-contrast grating and weakened the SSVEP signal associated with the high-contrast grating. As this would result in a smaller difference between these signals, and therefore, a lower amplitude marginal SSVEP compared to the neutral cue condition, this could explain the effect we observed.

It would also be expected that a valid cue should have the converse effect, producing a stronger marginal SSVEP compared to the neutral cue condition. This pattern was not borne out in the data. However, there is reason to believe that the cued and neutral conditions do not only differ in terms of the predictive information conferred to the participant. Unlike in the cued conditions, participants were given no prior indication of the relative probability of the upcoming target in the neutral cue condition, meaning that they were required to rely solely on careful accumulation of the presented evidence with no preparatory strategic adjustments on a demanding task, designed to elicit only 70% accuracy. Indeed, analysis of pre-evidence alpha activity over occipito-parietal sites provided evidence of enhanced stimulus engagement in the neutral cue condition compared to the cued conditions. Variations in pre-stimulus alpha-band power have been shown to correlate with fluctuations in performance in a variety of perceptual tasks (e.g. Babiloni et al., 2006; Ergenoglu et al., 2004; Hanslmayr et al., 2007; Linkenkaer-Hansen et al., 2004; O’Connell et al., 2009; van Dijk et al., 2008), a phenomenon thought to arise from the preferential encoding of task-relevant information (Jensen and Mazaheri, 2010; Van Diepen et al., 2019). Together, this suggests that participants may have been more closely engaged with the stimulus on neutral cue trials and, as attention has previously been shown to boost the SSVEP (e.g. Morgan et al., 1996; Müller et al., 2006), this may explain the similar SSVEP waveforms in the valid and neutral cue conditions. This also raises the possibility that a valid cue may actually boost the SSVEP rather than, or as well as, the invalid cue suppressing the SSVEP since it appears that participants were able to achieve the same marginal SSVEP amplitude on valid as neutral cue trials with less effort/attention. This could be investigated in future research by using cues with scaled predictive probabilities (e.g. 60%, 70%, and 80%) made explicit to the participant. The direction of the modulation could be characterised by assessing changes in the response of both the target and non-target SSVEPs following cues with different probabilities.

Modelling studies have indicated that starting-point biases are the optimal way to incorporate prior knowledge into the decision process (Bogacz et al., 2006; van Ravenzwaaij et al., 2012) and are sufficient to account for prior probability effects on behaviour without the need to invoke modulations of the evidence itself (Leite and Ratcliff, 2011; Mulder et al., 2012; Ratcliff and McKoon, 2008). However, the emphasis on parsimony in formal model comparisons has tended to promote models that can capture experimental manipulations of behaviour with a single dominant parameter adjustment (Purcell and Palmeri, 2017; Voss et al., 2004), like a starting-point bias. Our model-free neurophysiological analyses reveal that the strong biases in motor preparation, which participants exhibited early on in testing, were accompanied by more subtle sensory encoding modulations, which only emerged with extensive task exposure (circa 4 hr). Participants’ scores remained constant across task exposure, even as difficulty titration reduced the available physical evidence. So where one might have predicted that participants would learn to decrease their reliance on suboptimal strategies as task performance improves, our results point to the opposite trend. We suggest that participants may have been able to stabilise their performance across task exposure, despite reductions in the available sensory evidence, by incorporating the small sensory modulation we detected in the SSVEP. This would suggest that the decision process may not operate precisely as the models used in theoretical work describe. Instead, our study tentatively supports a small number of modelling investigations that have challenged the solitary role of starting point bias, implicating a drift bias (i.e. a modulation of the evidence before or upon entry to the decision variable) as an additional source of prior probability effects in perceptual decisions (Dunovan et al., 2014; Hanks et al., 2011; Kelly et al., 2021; van Ravenzwaaij et al., 2012; Wyart et al., 2012) and indicates that these drift biases could, at least partly, originate at the sensory level. However, this link could only be firmly established with modelling in a future study.

It has been suggested that sensory-level effects may reflect attentional processes or post-decisional relaying of the decision state to sensory regions, without fundamentally changing the encoding of the sensory evidence (Simon et al., 2019). Indeed, the dissociation of expectation and attention has been a Gordian knot in expectation research, particularly due to the prevalence of cueing paradigms (Aitchison and Lengyel, 2017; Alink and Blank, 2021; Rungratsameetaweemana and Serences, 2019; Summerfield and Egner, 2009). Several authors have pointed out that when prior probabilities are manipulated using cues, subjects are provided with both probabilistic information about what is likely to appear and information about the relevance of stimulus features for the task being performed, conflating expectation and attention (Bang and Rahnev, 2017; Rungratsameetaweemana and Serences, 2019; Simon et al., 2019). Indeed, shifts in feature-based attention have been found to induce feature-specific modulations of stimulus-evoked BOLD activity in early visual areas (e.g. Kamitani and Tong, 2005; Serences and Boynton, 2007). An advantage of the use of overlaid grating stimuli in the present paradigm was that there was no obvious advantage to shifting attention to a particular region of space to better identify one of the stimulus alternatives. In addition, both gratings were equally relevant to the choice, regardless of the correct alternative since participants were asked to compare their relative contrast. Nevertheless, the possibility that participants may have preferentially attended to the grating that was expected to appear at higher contrast cannot be excluded here. If an invalid cue led participants to pay closer attention to the non-target orientation, this may have degraded the differential representation of the contrast evidence. It is possible that the reason attention did not significantly enhance the marginal SSVEP when valid cues were presented was because the physical evidence had already guided attention to the target grating on neutral and valid cue trials. However, even if the SSVEP modulations reported here arise from feature-based attention rather than a sensory-prior, they would still constitute a modulation of the sensory evidence feeding the decision process based on prior probabilities.

An implication of using cues that predict not just the upcoming stimulus, but the most likely response, is that it becomes difficult to determine if the preparatory shifts in mu-beta (MB) activity that we observed reflect adjustments directly influencing the perceptual interpretation of the stimulus or simply preparation of the more probable action. When perceptual decisions are explicitly tied to particular modes of response, the decision state can be read from activity in motor regions associated with the preparation of that kind of action (e.g. de Lafuente et al., 2015; Ding and Gold, 2012; Shadlen and Newsome, 2001; Romo et al., 2004), but these modules appear to be part of a constellation of decision-related areas that are flexibly recruited based on the response modality (e.g. Filimon et al., 2013). When participants cannot prepare a response in advance or no response is required, MB no longer traces decision formation (Twomey et al., 2016), but an abstract decision process is still readily detectable (e.g. O’Connell et al., 2012), and modelling work suggests that drift biases and starting point biases continue to influence prior-informed decision making (Thomas et al., 2022; Yon et al., 2021). While the design of the present study does not allow us to offer further insight about whether the MB effects we observed were inherited from strategic adjustments at this abstract level of the decision process, we hope to conduct investigations in the future that better dissect the distinct components of prior-informed decisions to address this question.

Several other issues remain unaddressed by the present study. It is not clear to what extent the sensory effects may be influenced by features of the task design (e.g. speeded responses under a strict deadline) and if these sensory effects would generalise to many kinds of perceptual decision-making tasks or whether they are particular to contrast discrimination. This is an important area for further research as some of the studies that have reported no evidence of sensory modulations by expectation have used motion discrimination (Rao et al., 2012), orientation discrimination (Rungratsameetaweemana et al., 2018), or expanded judgement tasks (Bang and Rahnev, 2017). Additionally, in the present study, a pulse of evidence could be presented within the first 650 ms of a trial. To minimise contamination of the SSVEP dynamics with these evidence pulses, the effect of the cue was measured in the window 680–975ms, so it is not clear whether the cue also influenced the SSVEP in the earliest stages of evidence presentation. Some have suggested that expectation-based modulations may only influence the encoding of a stimulus after an initial volley of sensory processing (Aitken et al., 2020; Alilović et al., 2019; Press et al., 2020). Although the temporal resolution of the SSVEP is limited by the window of activity needed for Fourier analysis, future work could provide greater insight on when the cue effect emerged by removing the pulses from the paradigm. While our goal was to assess gradual trends in the emergence of these effects with task exposure, future studies could also investigate the possibility that this process is not linear. For example, participants may respond to the emergence of a sensory modulation with discrete strategic adjustments.

In conclusion, we used a contrast-sensitive electrophysiological signature of sensory activity to determine if prior probability exerted an influence on the encoding of sensory evidence feeding the decision process. Probabilistic cues were found to modulate the amplitude of the SSVEP, such that the representation of the marginal contrast of unexpected stimuli was relatively diminished. In addition, the effect only emerged over the course of prolonged exposure to the task. A modulation of the original encoding of the sensory evidence would have knock-on effects at each stage of the decision process. This suggests that, in addition to the preparatory strategic adjustments that are characteristic of expectation-based decision-making, prior probability may also be instantiated in the dynamics of the ongoing deliberation. This is consistent with recent work which implicates the subtle orchestration of several distinct parameter adjustments in prior-informed decisions (Kelly et al., 2021). We suggest that this effect may represent a candidate mechanism for a sensory-level contribution to this dynamic.

The datasets required to replicate the results and reproduce the figures in the manuscript are freely available, along with the task code, in the project's Open Science Framework repository (Walsh et al., 2023).

1. 1. Walsh K
   2. McGovern D
   3. Dully J
   4. Kelly S
   5. O'Connell R

   (2024) Open Science Framework

   ID 10.17605/OSF.IO/B92WM. Prior probability cues bias sensory encoding with increasing task exposure.

   https://osf.io/b92wm/

1. 1. Aitchison L
   2. Lengyel M

   (2017) With or without you: predictive coding and Bayesian inference in the brain

   Current Opinion in Neurobiology 46:219–227.

   https://doi.org/10.1016/j.conb.2017.08.010

   • PubMed
   • Google Scholar
2. 1. Aitken F
   2. Turner G
   3. Kok P

   (2020) Prior expectations of motion direction modulate early sensory processing

   The Journal of Neuroscience 40:6389–6397.

   https://doi.org/10.1523/JNEUROSCI.0537-20.2020

   • PubMed
   • Google Scholar
3. 1. Albright TD

   (2012) On the perception of probable things: neural substrates of associative memory, imagery, and perception

   Neuron 74:227–245.

   https://doi.org/10.1016/j.neuron.2012.04.001

   • PubMed
   • Google Scholar
4. 1. Alilović J
   2. Timmermans B
   3. Reteig LC
   4. van Gaal S
   5. Slagter HA

   (2019) No evidence that predictions and attention modulate the first feedforward sweep of cortical information processing

   Cerebral Cortex 29:2261–2278.

   https://doi.org/10.1093/cercor/bhz038

   • PubMed
   • Google Scholar
5. 1. Alink A
   2. Blank H

   (2021) Can expectation suppression be explained by reduced attention to predictable stimuli?

   NeuroImage 231:117824.

   https://doi.org/10.1016/j.neuroimage.2021.117824

   • PubMed
   • Google Scholar
6. 1. Baayen RH
   2. Davidson DJ
   3. Bates DM

   (2008) Mixed-effects modeling with crossed random effects for subjects and items

   Journal of Memory and Language 59:390–412.

   https://doi.org/10.1016/j.jml.2007.12.005

   • Google Scholar
7. 1. Babiloni C
   2. Vecchio F
   3. Bultrini A
   4. Luca Romani G
   5. Rossini PM

   (2006) Pre- and poststimulus alpha rhythms are related to conscious visual perception: a high-resolution EEG study

   Cerebral Cortex 16:1690–1700.

   https://doi.org/10.1093/cercor/bhj104

   • PubMed
   • Google Scholar
8. 1. Baker DH
   2. Vilidaite G
   3. Lygo FA
   4. Smith AK
   5. Flack TR
   6. Gouws AD
   7. Andrews TJ

   (2021) Power contours: optimising sample size and precision in experimental psychology and human neuroscience

   Psychological Methods 26:295–314.

   https://doi.org/10.1037/met0000337

   • PubMed
   • Google Scholar
9. 1. Bang JW
   2. Rahnev D

   (2017) Stimulus expectation alters decision criterion but not sensory signal in perceptual decision making

   Scientific Reports 7:17072.

   https://doi.org/10.1038/s41598-017-16885-2

   • PubMed
   • Google Scholar
10. 1. Blom T
    2. Feuerriegel D
    3. Johnson P
    4. Bode S
    5. Hogendoorn H

    (2020) Predictions drive neural representations of visual events ahead of incoming sensory information

    PNAS 117:7510–7515.

    https://doi.org/10.1073/pnas.1917777117

    • PubMed
    • Google Scholar
11. 1. Bogacz R
    2. Brown E
    3. Moehlis J
    4. Holmes P
    5. Cohen JD

    (2006) The physics of optimal decision making: a formal analysis of models of performance in two-alternative forced-choice tasks

    Psychological Review 113:700–765.

    https://doi.org/10.1037/0033-295X.113.4.700

    • PubMed
    • Google Scholar
12. 1. Cohen MX

    (2017) Where does EEG come from and what does it mean?

    Trends in Neurosciences 40:208–218.

    https://doi.org/10.1016/j.tins.2017.02.004

    • PubMed
    • Google Scholar
13. 1. Crick F
    2. Koch C

    (1998) Constraints on cortical and thalamic projections: the no-strong-loops hypothesis

    Nature 391:245–250.

    https://doi.org/10.1038/34584

    • PubMed
    • Google Scholar
14. 1. de Lafuente V
    2. Jazayeri M
    3. Shadlen MN

    (2015) Representation of accumulating evidence for a decision in two parietal areas

    The Journal of Neuroscience 35:4306–4318.

    https://doi.org/10.1523/JNEUROSCI.2451-14.2015

    • PubMed
    • Google Scholar
15. 1. de Lange FP
    2. Rahnev DA
    3. Donner TH
    4. Lau H

    (2013) Prestimulus oscillatory activity over motor cortex reflects perceptual expectations

    The Journal of Neuroscience 33:1400–1410.

    https://doi.org/10.1523/JNEUROSCI.1094-12.2013

    • PubMed
    • Google Scholar
16. 1. de Lange FP
    2. Heilbron M
    3. Kok P

    (2018) How do expectations shape perception?

    Trends in Cognitive Sciences 22:764–779.

    https://doi.org/10.1016/j.tics.2018.06.002

    • PubMed
    • Google Scholar
17. 1. Delorme A
    2. Makeig S

    (2004) EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis

    Journal of Neuroscience Methods 134:9–21.

    https://doi.org/10.1016/j.jneumeth.2003.10.009

    • PubMed
    • Google Scholar
18. 1. Deneve S

    (2012) Making decisions with unknown sensory reliability

    Frontiers in Neuroscience 6:75.

    https://doi.org/10.3389/fnins.2012.00075

    • PubMed
    • Google Scholar
19. 1. den Ouden C
    2. Zhou A
    3. Mepani V
    4. Kovács G
    5. Vogels R
    6. Feuerriegel D

    (2023) Stimulus expectations do not modulate visual event-related potentials in probabilistic cueing designs

    NeuroImage 280:120347.

    https://doi.org/10.1016/j.neuroimage.2023.120347

    • PubMed
    • Google Scholar
20. 1. Devine CA
    2. Gaffney C
    3. Loughnane GM
    4. Kelly SP
    5. O’Connell RG

    (2019) The role of premature evidence accumulation in making difficult perceptual decisions under temporal uncertainty

    eLife 8:e48526.

    https://doi.org/10.7554/eLife.48526

    • PubMed
    • Google Scholar
21. 1. Ding L
    2. Gold JI

    (2012) Neural correlates of perceptual decision making before, during, and after decision commitment in monkey frontal eye field

    Cerebral Cortex 22:1052–1067.

    https://doi.org/10.1093/cercor/bhr178

    • PubMed
    • Google Scholar
22. 1. Di Russo F
    2. Pitzalis S
    3. Aprile T
    4. Spitoni G
    5. Patria F
    6. Stella A
    7. Spinelli D
    8. Hillyard SA

    (2007) Spatiotemporal analysis of the cortical sources of the steady-state visual evoked potential

    Human Brain Mapping 28:323–334.

    https://doi.org/10.1002/hbm.20276

    • PubMed
    • Google Scholar
23. 1. Donner TH
    2. Siegel M
    3. Fries P
    4. Engel AK

    (2009) Buildup of choice-predictive activity in human motor cortex during perceptual decision making

    Current Biology 19:1581–1585.

    https://doi.org/10.1016/j.cub.2009.07.066

    • PubMed
    • Google Scholar
24. 1. Dunovan KE
    2. Tremel JJ
    3. Wheeler ME

    (2014) Prior probability and feature predictability interactively bias perceptual decisions

    Neuropsychologia 61:210–221.

    https://doi.org/10.1016/j.neuropsychologia.2014.06.024

    • PubMed
    • Google Scholar
25. 1. Egner T
    2. Monti JM
    3. Summerfield C

    (2010) Expectation and surprise determine neural population responses in the ventral visual stream

    The Journal of Neuroscience 30:16601–16608.

    https://doi.org/10.1523/JNEUROSCI.2770-10.2010

    • PubMed
    • Google Scholar
26. 1. Ekman M
    2. Kok P
    3. de Lange FP

    (2017) Time-compressed preplay of anticipated events in human primary visual cortex

    Nature Communications 8:15276.

    https://doi.org/10.1038/ncomms15276

    • PubMed
    • Google Scholar
27. 1. Ergenoglu T
    2. Demiralp T
    3. Bayraktaroglu Z
    4. Ergen M
    5. Beydagi H
    6. Uresin Y

    (2004) Alpha rhythm of the EEG modulates visual detection performance in humans

    Brain Research. Cognitive Brain Research 20:376–383.

    https://doi.org/10.1016/j.cogbrainres.2004.03.009

    • PubMed
    • Google Scholar
28. 1. Esterman M
    2. Yantis S

    (2010) Perceptual expectation evokes category-selective cortical activity

    Cerebral Cortex 20:1245–1253.

    https://doi.org/10.1093/cercor/bhp188

    • PubMed
    • Google Scholar
29. 1. Feuerriegel D
    2. Churches O
    3. Coussens S
    4. Keage HAD

    (2018) Evidence for spatiotemporally distinct effects of image repetition and perceptual expectations as measured by event-related potentials

    NeuroImage 169:94–105.

    https://doi.org/10.1016/j.neuroimage.2017.12.029

    • PubMed
    • Google Scholar
30. 1. Feuerriegel D
    2. Blom T
    3. Hogendoorn H

    (2021a) Predictive activation of sensory representations as a source of evidence in perceptual decision-making

    Cortex; a Journal Devoted to the Study of the Nervous System and Behavior 136:140–146.

    https://doi.org/10.1016/j.cortex.2020.12.008

    • PubMed
    • Google Scholar
31. 1. Feuerriegel D
    2. Vogels R
    3. Kovács G

    (2021b) Evaluating the evidence for expectation suppression in the visual system

    Neuroscience and Biobehavioral Reviews 126:368–381.

    https://doi.org/10.1016/j.neubiorev.2021.04.002

    • PubMed
    • Google Scholar
32. 1. Feuerriegel D
    2. Yook J
    3. Quek GL
    4. Hogendoorn H
    5. Bode S

    (2021c) Visual mismatch responses index surprise signalling but not expectation suppression

    Cortex; a Journal Devoted to the Study of the Nervous System and Behavior 134:16–29.

    https://doi.org/10.1016/j.cortex.2020.10.006

    • PubMed
    • Google Scholar
33. 1. Filimon F
    2. Philiastides MG
    3. Nelson JD
    4. Kloosterman NA
    5. Heekeren HR

    (2013) How embodied is perceptual decision making? Evidence for separate processing of perceptual and motor decisions

    The Journal of Neuroscience 33:2121–2136.

    https://doi.org/10.1523/JNEUROSCI.2334-12.2013

    • PubMed
    • Google Scholar
34. Preprint

    1. Findling C
    2. Hubert F
    3. Acerbi L
    4. Benson B
    5. Benson J
    6. Birman D
    7. Bonacchi N
    8. Carandini M
    9. Catarino JA
    10. Chapuis GA
    11. Churchland AK
    12. Dan Y
    13. DeWitt EE
    14. Engel TA
    15. Fabbri M
    16. Faulkner M
    17. Fiete IR
    18. Freitas-Silva L
    19. Gerçek B
    20. Harris KD
    21. Häusser M
    22. Hofer SB
    23. Hu F
    24. Huntenburg JM
    25. Khanal A
    26. Krasniak C
    27. Langdon C
    28. Latham PE
    29. P Lau PY
    30. Mainen Z
    31. Meijer GT
    32. Miska NJ
    33. Mrsic-Flogel TD
    34. Noel JP
    35. Nylund K
    36. Pan-Vazquez A
    37. Paninski L
    38. Pillow J
    39. Rossant C
    40. Roth N
    41. Schaeffer R
    42. Schartner M
    43. Shi Y
    44. Socha KZ
    45. Steinmetz NA
    46. Svoboda K
    47. Tessereau C
    48. Urai AE
    49. Wells MJ
    50. West SJ
    51. Whiteway MR
    52. Winter O
    53. Witten IB
    54. Zador A
    55. Dayan P
    56. Pouget A
    57. International Brain Laboratory

    (2023) Brain-wide representations of prior information in mouse decision-making

    bioRxiv.

    https://doi.org/10.1101/2023.07.04.547684

    • Google Scholar
35. 1. Fiser A
    2. Mahringer D
    3. Oyibo HK
    4. Petersen AV
    5. Leinweber M
    6. Keller GB

    (2016) Experience-dependent spatial expectations in mouse visual cortex

    Nature Neuroscience 19:1658–1664.

    https://doi.org/10.1038/nn.4385

    • PubMed
    • Google Scholar
36. 1. Friston K

    (2005) A theory of cortical responses

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 360:815–836.

    https://doi.org/10.1098/rstb.2005.1622

    • PubMed
    • Google Scholar
37. 1. Friston K

    (2009) The free-energy principle: a rough guide to the brain?

    Trends in Cognitive Sciences 13:293–301.

    https://doi.org/10.1016/j.tics.2009.04.005

    • PubMed
    • Google Scholar
38. 1. Friston K

    (2010) The free-energy principle: a unified brain theory?

    Nature Reviews. Neuroscience 11:127–138.

    https://doi.org/10.1038/nrn2787

    • PubMed
    • Google Scholar
39. 1. Gavornik JP
    2. Bear MF

    (2014) Learned spatiotemporal sequence recognition and prediction in primary visual cortex

    Nature Neuroscience 17:732–737.

    https://doi.org/10.1038/nn.3683

    • PubMed
    • Google Scholar
40. 1. Gold JI
    2. Stocker AA

    (2017) Visual decision-making in an uncertain and dynamic world

    Annual Review of Vision Science 3:227–250.

    https://doi.org/10.1146/annurev-vision-111815-114511

    • PubMed
    • Google Scholar
41. Preprint

    1. Grogan JP
    2. Rys W
    3. Kelly SP
    4. O’Connell RG

    (2023) Confidence is predicted by pre- and post-choice decision signal dynamics

    bioRxiv.

    https://doi.org/10.1101/2023.01.19.524702

    • Google Scholar
42. 1. Grotheer M
    2. Kovács G

    (2015) The relationship between stimulus repetitions and fulfilled expectations

    Neuropsychologia 67:175–182.

    https://doi.org/10.1016/j.neuropsychologia.2014.12.017

    • PubMed
    • Google Scholar
43. 1. Hall MG
    2. Mattingley JB
    3. Dux PE

    (2018) Electrophysiological correlates of incidentally learned expectations in human vision

    Journal of Neurophysiology 119:1461–1470.

    https://doi.org/10.1152/jn.00733.2017

    • PubMed
    • Google Scholar
44. 1. Hanks TD
    2. Mazurek ME
    3. Kiani R
    4. Hopp E
    5. Shadlen MN

    (2011) Elapsed decision time affects the weighting of prior probability in a perceptual decision task

    The Journal of Neuroscience 31:6339–6352.

    https://doi.org/10.1523/JNEUROSCI.5613-10.2011

    • PubMed
    • Google Scholar
45. 1. Hanslmayr S
    2. Aslan A
    3. Staudigl T
    4. Klimesch W
    5. Herrmann CS
    6. Bäuml KH

    (2007) Prestimulus oscillations predict visual perception performance between and within subjects

    NeuroImage 37:1465–1473.

    https://doi.org/10.1016/j.neuroimage.2007.07.011

    • PubMed
    • Google Scholar
46. 1. Heeger DJ

    (2017) Theory of cortical function

    PNAS 114:1773–1782.

    https://doi.org/10.1073/pnas.1619788114

    • PubMed
    • Google Scholar
47. 1. Heilbron M
    2. Chait M

    (2018) Great expectations: is there evidence for predictive coding in auditory cortex?

    Neuroscience 389:54–73.

    https://doi.org/10.1016/j.neuroscience.2017.07.061

    • PubMed
    • Google Scholar
48. 1. Jensen O
    2. Mazaheri A

    (2010) Shaping functional architecture by oscillatory alpha activity: gating by inhibition

    Frontiers in Human Neuroscience 4:186.

    https://doi.org/10.3389/fnhum.2010.00186

    • PubMed
    • Google Scholar
49. 1. Kamitani Y
    2. Tong F

    (2005) Decoding the visual and subjective contents of the human brain

    Nature Neuroscience 8:679–685.

    https://doi.org/10.1038/nn1444

    • PubMed
    • Google Scholar
50. 1. Kaposvari P
    2. Kumar S
    3. Vogels R

    (2018) Statistical learning signals in macaque inferior temporal cortex

    Cerebral Cortex 28:250–266.

    https://doi.org/10.1093/cercor/bhw374

    • PubMed
    • Google Scholar
51. 1. Kayser J
    2. Tenke CE

    (2015) On the benefits of using surface Laplacian (current source density) methodology in electrophysiology

    International Journal of Psychophysiology 97:171–173.

    https://doi.org/10.1016/j.ijpsycho.2015.06.001

    • PubMed
    • Google Scholar
52. 1. Kelly SP
    2. Lalor EC
    3. Reilly RB
    4. Foxe JJ

    (2006) Increases in alpha oscillatory power reflect an active retinotopic mechanism for distracter suppression during sustained visuospatial attention

    Journal of Neurophysiology 95:3844–3851.

    https://doi.org/10.1152/jn.01234.2005

    • PubMed
    • Google Scholar
53. 1. Kelly SP
    2. Gomez-Ramirez M
    3. Foxe JJ

    (2009) The strength of anticipatory spatial biasing predicts target discrimination at attended locations: a high-density EEG study

    The European Journal of Neuroscience 30:2224–2234.

    https://doi.org/10.1111/j.1460-9568.2009.06980.x

    • PubMed
    • Google Scholar
54. 1. Kelly SP
    2. O’Connell RG

    (2013) Internal and external influences on the rate of sensory evidence accumulation in the human brain

    The Journal of Neuroscience 33:19434–19441.

    https://doi.org/10.1523/JNEUROSCI.3355-13.2013

    • PubMed
    • Google Scholar
55. 1. Kelly SP
    2. Corbett EA
    3. O’Connell RG

    (2021) Neurocomputational mechanisms of prior-informed perceptual decision-making in humans

    Nature Human Behaviour 5:467–481.

    https://doi.org/10.1038/s41562-020-00967-9

    • PubMed
    • Google Scholar
56. 1. Kim YJ
    2. Grabowecky M
    3. Paller KA
    4. Muthu K
    5. Suzuki S

    (2007) Attention induces synchronization-based response gain in steady-state visual evoked potentials

    Nature Neuroscience 10:117–125.

    https://doi.org/10.1038/nn1821

    • PubMed
    • Google Scholar
57. 1. Kleiner M
    2. Brainard D
    3. Pelli D

    (2007)

    What’s new in Psychtoolbox-3?

    Perception 36:1–16.

    • Google Scholar
58. 1. Koch C
    2. Poggio T

    (1999) Predicting the visual world: silence is golden

    Nature Neuroscience 2:9–10.

    https://doi.org/10.1038/4511

    • PubMed
    • Google Scholar
59. 1. Kok P
    2. Jehee JFM
    3. de Lange FP

    (2012) Less is more: expectation sharpens representations in the primary visual cortex

    Neuron 75:265–270.

    https://doi.org/10.1016/j.neuron.2012.04.034

    • PubMed
    • Google Scholar
60. 1. Kok P
    2. Brouwer GJ
    3. van Gerven MAJ
    4. de Lange FP

    (2013) Prior expectations bias sensory representations in visual cortex

    The Journal of Neuroscience 33:16275–16284.

    https://doi.org/10.1523/JNEUROSCI.0742-13.2013

    • PubMed
    • Google Scholar
61. 1. Kok P
    2. Failing MF
    3. de Lange FP

    (2014) Prior expectations evoke stimulus templates in the primary visual cortex

    Journal of Cognitive Neuroscience 26:1546–1554.

    https://doi.org/10.1162/jocn_a_00562

    • PubMed
    • Google Scholar
62. 1. Kok P
    2. Mostert P
    3. de Lange FP

    (2017) Prior expectations induce prestimulus sensory templates

    PNAS 114:10473–10478.

    https://doi.org/10.1073/pnas.1705652114

    • PubMed
    • Google Scholar
63. 1. Kuznetsova A
    2. Brockhoff PB
    3. Christensen RHB

    (2017) lmerTest package: tests in linear mixed effects models

    Journal of Statistical Software 82:1–26.

    https://doi.org/10.18637/jss.v082.i13

    • Google Scholar
64. 1. Laufs H
    2. Krakow K
    3. Sterzer P
    4. Eger E
    5. Beyerle A
    6. Salek-Haddadi A
    7. Kleinschmidt A

    (2003) Electroencephalographic signatures of attentional and cognitive default modes in spontaneous brain activity fluctuations at rest

    PNAS 100:11053–11058.

    https://doi.org/10.1073/pnas.1831638100

    • PubMed
    • Google Scholar
65. 1. Laufs H
    2. Holt JL
    3. Elfont R
    4. Krams M
    5. Paul JS
    6. Krakow K
    7. Kleinschmidt A

    (2006) Where the BOLD signal goes when alpha EEG leaves

    NeuroImage 31:1408–1418.

    https://doi.org/10.1016/j.neuroimage.2006.02.002

    • PubMed
    • Google Scholar
66. 1. Lauritzen TZ
    2. Ales JM
    3. Wade AR

    (2010) The effects of visuospatial attention measured across visual cortex using source-imaged, steady-state EEG

    Journal of Vision 10:39.

    https://doi.org/10.1167/10.14.39

    • PubMed
    • Google Scholar
67. 1. Leite FP
    2. Ratcliff R

    (2011) What cognitive processes drive response biases? A diffusion model analysis

    Judgment and Decision Making 6:651–687.

    https://doi.org/10.1017/S1930297500002680

    • Google Scholar
68. 1. Linkenkaer-Hansen K
    2. Nikulin VV
    3. Palva S
    4. Ilmoniemi RJ
    5. Palva JM

    (2004) Prestimulus oscillations enhance psychophysical performance in humans

    The Journal of Neuroscience 24:10186–10190.

    https://doi.org/10.1523/JNEUROSCI.2584-04.2004

    • PubMed
    • Google Scholar
69. 1. Logothetis NK

    (2008) What we can do and what we cannot do with fMRI

    Nature 453:869–878.

    https://doi.org/10.1038/nature06976

    • PubMed
    • Google Scholar
70. 1. Luke SG

    (2017) Evaluating significance in linear mixed-effects models in R

    Behavior Research Methods 49:1494–1502.

    https://doi.org/10.3758/s13428-016-0809-y

    • PubMed
    • Google Scholar
71. 1. Mantini D
    2. Perrucci MG
    3. Del Gratta C
    4. Romani GL
    5. Corbetta M

    (2007) Electrophysiological signatures of resting state networks in the human brain

    PNAS 104:13170–13175.

    https://doi.org/10.1073/pnas.0700668104

    • PubMed
    • Google Scholar
72. 1. Meyer T
    2. Olson CR

    (2011) Statistical learning of visual transitions in monkey inferotemporal cortex

    PNAS 108:19401–19406.

    https://doi.org/10.1073/pnas.1112895108

    • PubMed
    • Google Scholar
73. 1. Meyer T
    2. Ramachandran S
    3. Olson CR

    (2014) Statistical learning of serial visual transitions by neurons in monkey inferotemporal cortex

    The Journal of Neuroscience 34:9332–9337.

    https://doi.org/10.1523/JNEUROSCI.1215-14.2014

    • PubMed
    • Google Scholar
74. 1. Morgan ST
    2. Hansen JC
    3. Hillyard SA

    (1996) Selective attention to stimulus location modulates the steady-state visual evoked potential

    PNAS 93:4770–4774.

    https://doi.org/10.1073/pnas.93.10.4770

    • PubMed
    • Google Scholar
75. 1. Mulder MJ
    2. Wagenmakers EJ
    3. Ratcliff R
    4. Boekel W
    5. Forstmann BU

    (2012) Bias in the brain: A diffusion model analysis of prior probability and potential payoff

    The Journal of Neuroscience 32:2335–2343.

    https://doi.org/10.1523/JNEUROSCI.4156-11.2012

    • PubMed
    • Google Scholar
76. 1. Mulder MJ
    2. van Maanen L
    3. Forstmann BU

    (2014) Perceptual decision neurosciences - a model-based review

    Neuroscience 277:872–884.

    https://doi.org/10.1016/j.neuroscience.2014.07.031

    • PubMed
    • Google Scholar
77. 1. Müller MM
    2. Andersen S
    3. Trujillo NJ
    4. Valdés-Sosa P
    5. Malinowski P
    6. Hillyard SA

    (2006) Feature-selective attention enhances color signals in early visual areas of the human brain

    PNAS 103:14250–14254.

    https://doi.org/10.1073/pnas.0606668103

    • PubMed
    • Google Scholar
78. 1. Norcia AM
    2. Appelbaum LG
    3. Ales JM
    4. Cottereau BR
    5. Rossion B

    (2015) The steady-state visual evoked potential in vision research: A review

    Journal of Vision 15:4.

    https://doi.org/10.1167/15.6.4

    • PubMed
    • Google Scholar
79. 1. O’Connell RG
    2. Dockree PM
    3. Robertson IH
    4. Bellgrove MA
    5. Foxe JJ
    6. Kelly SP

    (2009) Uncovering the neural signature of lapsing attention: electrophysiological signals predict errors up to 20 s before they occur

    The Journal of Neuroscience 29:8604–8611.

    https://doi.org/10.1523/JNEUROSCI.5967-08.2009

    • PubMed
    • Google Scholar
80. 1. O’Connell RG
    2. Dockree PM
    3. Kelly SP

    (2012) A supramodal accumulation-to-bound signal that determines perceptual decisions in humans

    Nature Neuroscience 15:1729–1735.

    https://doi.org/10.1038/nn.3248

    • PubMed
    • Google Scholar
81. 1. Pouget A
    2. Beck JM
    3. Ma WJ
    4. Latham PE

    (2013) Probabilistic brains: knowns and unknowns

    Nature Neuroscience 16:1170–1178.

    https://doi.org/10.1038/nn.3495

    • PubMed
    • Google Scholar
82. 1. Press C
    2. Kok P
    3. Yon D

    (2020) The Perceptual Prediction Paradox

    Trends in Cognitive Sciences 24:13–24.

    https://doi.org/10.1016/j.tics.2019.11.003

    • PubMed
    • Google Scholar
83. 1. Purcell BA
    2. Palmeri TJ

    (2017) Relating accumulator model parameters and neural dynamics

    Journal of Mathematical Psychology 76:156–171.

    https://doi.org/10.1016/j.jmp.2016.07.001

    • PubMed
    • Google Scholar
84. 1. Puri AM
    2. Wojciulik E
    3. Ranganath C

    (2009) Category expectation modulates baseline and stimulus-evoked activity in human inferotemporal cortex

    Brain Research 1301:89–99.

    https://doi.org/10.1016/j.brainres.2009.08.085

    • PubMed
    • Google Scholar
85. 1. Pylyshyn Z

    (1999) Is vision continuous with cognition? The case for cognitive impenetrability of visual perception

    The Behavioral and Brain Sciences 22:341–365.

    https://doi.org/10.1017/s0140525x99002022

    • PubMed
    • Google Scholar
86. 1. Rahnev D
    2. Lau H
    3. de Lange FP

    (2011) Prior expectation modulates the interaction between sensory and prefrontal regions in the human brain

    The Journal of Neuroscience 31:10741–10748.

    https://doi.org/10.1523/JNEUROSCI.1478-11.2011

    • PubMed
    • Google Scholar
87. 1. Ramachandran S
    2. Meyer T
    3. Olson CR

    (2017) Prediction suppression and surprise enhancement in monkey inferotemporal cortex

    Journal of Neurophysiology 118:374–382.

    https://doi.org/10.1152/jn.00136.2017

    • PubMed
    • Google Scholar
88. 1. Rao V
    2. DeAngelis GC
    3. Snyder LH

    (2012) Neural correlates of prior expectations of motion in the lateral intraparietal and middle temporal areas

    The Journal of Neuroscience 32:10063–10074.

    https://doi.org/10.1523/JNEUROSCI.5948-11.2012

    • PubMed
    • Google Scholar
89. 1. Ratcliff R
    2. Smith PL

    (2004) A comparison of sequential sampling models for two-choice reaction time

    Psychological Review 111:333–367.

    https://doi.org/10.1037/0033-295X.111.2.333

    • PubMed
    • Google Scholar
90. 1. Ratcliff R
    2. McKoon G

    (2008) The diffusion decision model: theory and data for two-choice decision tasks

    Neural Computation 20:873–922.

    https://doi.org/10.1162/neco.2008.12-06-420

    • PubMed
    • Google Scholar
91. 1. Richter D
    2. Ekman M
    3. de Lange FP

    (2018) Suppressed sensory response to predictable object stimuli throughout the ventral visual stream

    The Journal of Neuroscience 38:7452–7461.

    https://doi.org/10.1523/JNEUROSCI.3421-17.2018

    • PubMed
    • Google Scholar
92. 1. Romei V
    2. Brodbeck V
    3. Michel C
    4. Amedi A
    5. Pascual-Leone A
    6. Thut G

    (2008) Spontaneous fluctuations in posterior alpha-band EEG activity reflect variability in excitability of human visual areas

    Cerebral Cortex 18:2010–2018.

    https://doi.org/10.1093/cercor/bhm229

    • PubMed
    • Google Scholar
93. 1. Romo R
    2. Hernández A
    3. Zainos A

    (2004) Neuronal correlates of a perceptual decision in ventral premotor cortex

    Neuron 41:165–173.

    https://doi.org/10.1016/S0896-6273(03)00817-1

    • Google Scholar
94. 1. Rungratsameetaweemana N
    2. Itthipuripat S
    3. Salazar A
    4. Serences JT

    (2018) Expectations do not alter early sensory processing during perceptual decision-making

    The Journal of Neuroscience 38:5632–5648.

    https://doi.org/10.1523/JNEUROSCI.3638-17.2018

    • PubMed
    • Google Scholar
95. 1. Rungratsameetaweemana N
    2. Serences JT

    (2019) Dissociating the impact of attention and expectation on early sensory processing

    Current Opinion in Psychology 29:181–186.

    https://doi.org/10.1016/j.copsyc.2019.03.014

    • PubMed
    • Google Scholar
96. 1. Sadaghiani S
    2. Scheeringa R
    3. Lehongre K
    4. Morillon B
    5. Giraud AL
    6. Kleinschmidt A

    (2010) Intrinsic connectivity networks, alpha oscillations, and tonic alertness: a simultaneous electroencephalography/functional magnetic resonance imaging study

    The Journal of Neuroscience 30:10243–10250.

    https://doi.org/10.1523/JNEUROSCI.1004-10.2010

    • PubMed
    • Google Scholar
97. 1. Scheeringa R
    2. Petersson KM
    3. Oostenveld R
    4. Norris DG
    5. Hagoort P
    6. Bastiaansen MCM

    (2009) Trial-by-trial coupling between EEG and BOLD identifies networks related to alpha and theta EEG power increases during working memory maintenance

    NeuroImage 44:1224–1238.

    https://doi.org/10.1016/j.neuroimage.2008.08.041

    • PubMed
    • Google Scholar
98. 1. Schlack A
    2. Albright TD

    (2007) Remembering visual motion: neural correlates of associative plasticity and motion recall in cortical area MT

    Neuron 53:881–890.

    https://doi.org/10.1016/j.neuron.2007.02.028

    • PubMed
    • Google Scholar
99. 1. Schwiedrzik CM
    2. Freiwald WA

    (2017) High-level prediction signals in a low-level area of the macaque face-processing hierarchy

    Neuron 96:89–97.

    https://doi.org/10.1016/j.neuron.2017.09.007

    • PubMed
    • Google Scholar
100. 1. Serences JT
     2. Boynton GM

     (2007) Feature-based attentional modulations in the absence of direct visual stimulation

     Neuron 55:301–312.

     https://doi.org/10.1016/j.neuron.2007.06.015

     • PubMed
     • Google Scholar
101. 1. Shadlen MN
     2. Newsome WT

     (2001) Neural basis of a perceptual decision in the parietal cortex (area LIP) of the rhesus monkey

     Journal of Neurophysiology 86:1916–1936.

     https://doi.org/10.1152/jn.2001.86.4.1916

     • PubMed
     • Google Scholar
102. 1. Simon AJ
     2. Schachtner JN
     3. Gallen CL

     (2019) Disentangling expectation from selective attention during perceptual decision making

     Journal of Neurophysiology 121:1977–1980.

     https://doi.org/10.1152/jn.00639.2018

     • PubMed
     • Google Scholar
103. 1. Smith PL
     2. Little DR

     (2018) Small is beautiful: In defense of the small-N design

     Psychonomic Bulletin & Review 25:2083–2101.

     https://doi.org/10.3758/s13423-018-1451-8

     • PubMed
     • Google Scholar
104. 1. Solomon SS
     2. Tang H
     3. Sussman E
     4. Kohn A

     (2021) Limited evidence for sensory prediction error responses in visual cortex of macaques and humans

     Cerebral Cortex 31:3136–3152.

     https://doi.org/10.1093/cercor/bhab014

     • PubMed
     • Google Scholar
105. 1. Stefanics G
     2. Kremláček J
     3. Czigler I

     (2014) Visual mismatch negativity: a predictive coding view

     Frontiers in Human Neuroscience 8:666.

     https://doi.org/10.3389/fnhum.2014.00666

     • PubMed
     • Google Scholar
106. 1. Steinemann NA
     2. O’Connell RG
     3. Kelly SP

     (2018) Decisions are expedited through multiple neural adjustments spanning the sensorimotor hierarchy

     Nature Communications 9:3627.

     https://doi.org/10.1038/s41467-018-06117-0

     • PubMed
     • Google Scholar
107. 1. Summerfield C
     2. Egner T

     (2009) Expectation (and attention) in visual cognition

     Trends in Cognitive Sciences 13:403–409.

     https://doi.org/10.1016/j.tics.2009.06.003

     • PubMed
     • Google Scholar
108. 1. Summerfield C
     2. de Lange FP

     (2014) Expectation in perceptual decision making: neural and computational mechanisms

     Nature Reviews. Neuroscience 15:745–756.

     https://doi.org/10.1038/nrn3838

     • PubMed
     • Google Scholar
109. 1. Tang MF
     2. Smout CA
     3. Arabzadeh E
     4. Mattingley JB

     (2018) Prediction error and repetition suppression have distinct effects on neural representations of visual information

     eLife 7:e33123.

     https://doi.org/10.7554/eLife.33123

     • PubMed
     • Google Scholar
110. 1. Teufel C
     2. Fletcher PC

     (2020) Forms of prediction in the nervous system

     Nature Reviews. Neuroscience 21:231–242.

     https://doi.org/10.1038/s41583-020-0275-5

     • PubMed
     • Google Scholar
111. 1. Thomas ER
     2. Yon D
     3. de Lange FP
     4. Press C

     (2022) Action Enhances Predicted Touch

     Psychological Science 33:48–59.

     https://doi.org/10.1177/09567976211017505

     • PubMed
     • Google Scholar
112. 1. Thut G
     2. Miniussi C

     (2009) New insights into rhythmic brain activity from TMS-EEG studies

     Trends in Cognitive Sciences 13:182–189.

     https://doi.org/10.1016/j.tics.2009.01.004

     • PubMed
     • Google Scholar
113. 1. Trapp S
     2. Lepsien J
     3. Kotz SA
     4. Bar M

     (2016) Prior probability modulates anticipatory activity in category-specific areas

     Cognitive, Affective & Behavioral Neuroscience 16:135–144.

     https://doi.org/10.3758/s13415-015-0373-4

     • PubMed
     • Google Scholar
114. 1. Twomey DM
     2. Kelly SP
     3. O’Connell RG

     (2016) Abstract and effector-selective decision signals exhibit qualitatively distinct dynamics before delayed perceptual reports

     The Journal of Neuroscience 36:7346–7352.

     https://doi.org/10.1523/JNEUROSCI.4162-15.2016

     • PubMed
     • Google Scholar
115. 1. Van Diepen RM
     2. Foxe JJ
     3. Mazaheri A

     (2019) The functional role of alpha-band activity in attentional processing: the current zeitgeist and future outlook

     Current Opinion in Psychology 29:229–238.

     https://doi.org/10.1016/j.copsyc.2019.03.015

     • PubMed
     • Google Scholar
116. 1. van Dijk H
     2. Schoffelen JM
     3. Oostenveld R
     4. Jensen O

     (2008) Prestimulus oscillatory activity in the alpha band predicts visual discrimination ability

     The Journal of Neuroscience 28:1816–1823.

     https://doi.org/10.1523/JNEUROSCI.1853-07.2008

     • PubMed
     • Google Scholar
117. 1. Vanegas MI
     2. Blangero A
     3. Kelly SP

     (2013) Exploiting individual primary visual cortex geometry to boost steady state visual evoked potentials

     Journal of Neural Engineering 10:036003.

     https://doi.org/10.1088/1741-2560/10/3/036003

     • PubMed
     • Google Scholar
118. 1. van Ravenzwaaij D
     2. Mulder MJ
     3. Tuerlinckx F
     4. Wagenmakers EJ

     (2012) Do the dynamics of prior information depend on task context? An analysis of optimal performance and an empirical test

     Frontiers in Psychology 3:132.

     https://doi.org/10.3389/fpsyg.2012.00132

     • PubMed
     • Google Scholar
119. 1. Voss A
     2. Rothermund K
     3. Voss J

     (2004) Interpreting the parameters of the diffusion model: an empirical validation

     Memory & Cognition 32:1206–1220.

     https://doi.org/10.3758/bf03196893

     • PubMed
     • Google Scholar
120. 1. Walsh KS
     2. McGovern DP
     3. Clark A
     4. O’Connell RG

     (2020) Evaluating the neurophysiological evidence for predictive processing as a model of perception

     Annals of the New York Academy of Sciences 1464:242–268.

     https://doi.org/10.1111/nyas.14321

     • PubMed
     • Google Scholar
121. Data

     1. Walsh K
     2. McGovern DP
     3. Dully J
     4. Kelly S
     5. O’Connell R

     (authors) (2023) Prior probability cues bias sensory encoding with increasing task exposure

     Open Science Framework.

     https://doi.org/10.17605/OSF.IO/B92WM
122. 1. Watson AB
     2. Pelli DG

     (1983) QUEST: A Bayesian adaptive psychometric method

     Perception & Psychophysics 33:113–120.

     https://doi.org/10.3758/bf03202828

     • PubMed
     • Google Scholar
123. 1. Wyart V
     2. Nobre AC
     3. Summerfield C

     (2012) Dissociable prior influences of signal probability and relevance on visual contrast sensitivity

     PNAS 109:3593–3598.

     https://doi.org/10.1073/pnas.1120118109

     • PubMed
     • Google Scholar
124. 1. Yon D
     2. Zainzinger V
     3. de Lange FP
     4. Eimer M
     5. Press C

     (2021) Action biases perceptual decisions toward expected outcomes

     Journal of Experimental Psychology. General 150:1225–1236.

     https://doi.org/10.1037/xge0000826

     • PubMed
     • Google Scholar

Author details

1. Kevin Walsh

   School of Psychological Sciences, Monash University, Melbourne, Australia

   Contribution

   Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing – review and editing

   For correspondence

   kevin.walsh1@monash.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3745-2073

2. David P McGovern

   School of Psychology, Dublin City University, Dublin, Ireland

   Contribution

   Conceptualization, Supervision, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5748-2827

3. Jessica Dully

   Institute of Psychiatry, Psychology & Neuroscience, King's College London, London, United Kingdom

   Contribution

   Conceptualization, Investigation, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0009-0007-2556-9403

4. Simon P Kelly

   1. School of Electrical Engineering, University College Dublin, Dublin, Ireland
   2. Trinity College Institute of Neuroscience, Trinity College Dublin, Dublin, Ireland

   Contribution

   Conceptualization, Supervision, Validation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9983-3595

5. Redmond G O'Connell

   1. Trinity College Institute of Neuroscience, Trinity College Dublin, Dublin, Ireland
   2. School of Psychology, Trinity College Dublin, Dublin, Ireland

   Contribution

   Conceptualization, Resources, Software, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Project administration, Writing – review and editing

   Competing interests

   Reviewing editor, eLife

   "This ORCID iD identifies the author of this article:" 0000-0001-6949-2793

• 771

  views

• 107

  downloads

• 3

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.