CoChat
Sign Up
Cited by 0Neural and Behavioral Psychology StudiesVisual perception and processing mechanismsFree Will and Agency

Evidence for unconscious priming using a Bayesian single-subject approach

Open access
Sánchez-Fuenzalida, Nicolás, van Gaal, Simon, van den Hurk, Zazie, Stein, Timo, Fahrenfort, Johannes J·2026·Neuroscience of Consciousness
DOI 10.1093/nc/niag052PMC PMC13492662
Open this paper in CoChat — free account

Abstract

Many studies have claimed the existence of unconscious priming, but a valid statistical procedure for substantiating such claims is often lacking. Absence of prime awareness is often inferred from non-significant p-values, and priming and awareness are typically tested by comparing two separate tests. In a classic study by Vorberg, D., Mattler, U., Heinecke, A., Schmidt, T., & Schwarzbach, J. (Different time courses for visual perception and action priming. Proceedings of the National Academy of Sciences 2003, 100(10), 6275–6280.), observers were slower to identify the direction of an arrow mask when it followed an arrow prime pointing in the opposite direction, despite being unable to identify the prime direction. Using this study as a starting point, we addressed several methodological shortcomings that are common in the field. First, we used a Bayesian methodological-statistical framework to test priming effects and prime awareness at the single-subject level across multiple sessions. Second, participants received extensive training in the prime awareness task to ensure they understood and executed the task as intended. Third, we directly compared prime awareness and priming. Using this approach, we found an indirect-task advantage consistent with unconscious priming at short prime-mask stimulus onset asynchronies (SOAs) in three out of six participants, under the assumption that the awareness task was at least as sensitive as the priming task. However, unlike in the original study by Vorberg and colleagues, evidence for unconscious priming was not consistently present across all SOAs, and awareness increased with SOA. Priming effects were highly consistent across participants, whereas awareness showed substantial inter-individual variability.

AI snapshot

Written from the paper’s full text.

TL;DR

Using a Bayesian single-subject framework across multiple sessions, the authors replicated a classic unconscious-priming paradigm and found an indirect-task advantage consistent with unconscious priming in some participants and SOAs, under the assumption that the awareness task was at least as sensitive as the priming task. Unlike Vorberg et al. (2003), prime awareness increased with SOA and varied substantially across individuals, while priming was highly consistent.

Key findings

  • In the initial control session, the original pattern was largely reproduced: prime discrimination showed anecdotal-to-very-strong evidence for chance at 12.5, 25, and 37.5 ms (BF10 = 0.035, 0.104, 0.772; d = 0.05, 0.13, 0.2), with above-chance performance at 50, 62.5, and 75 ms (BF10 = 4.59, 105, 33; d = 0.25, 0.32, 0.32); priming was reliable at 25–75 ms (e.g., 25 ms BF10 = 1015, d = 0.75; 75 ms BF10 = 1017, d = 1.86) but not at 12.5 ms (BF10 = 0.002, d = 0.11).
  • In the modified experiment, group-level prime discrimination was above chance at all SOAs and increased with SOA (12.5 ms BF10 = 107, d = 0.2; 75 ms BF10 = 1028, d = 1.1; 300 ms BF10 > 1040, d = 6.6), while priming was absent at 12.5 ms (BF10 = 0.009, d = 0.1) and present at 25–300 ms (e.g., 62.5 ms BF10 = 1013, d = 2; 300 ms BF10 = 1014, d = 1.3).
  • At the single-subject level, priming was highly consistent: all participants showed extreme evidence for no priming at 12.5 ms (BF10 < 0.01, d ≤ 0.2) and extreme evidence for priming at 25–75 ms (BF10 ≥ 1013, d ≥ 0.6). Prime awareness was variable; using the double t-test criterion, three participants met the original Vorberg-style criterion for unconscious priming in at least one SOA (participant 3 at 25, 37.5, and 50 ms; participant 4 at 25 ms; participant 5 at 25 and 37.5 ms).
  • Using median-split RT-to-accuracy recoding for a direct comparison, an indirect task advantage was found across all participants in at least one SOA; at the group level there was no difference at 12.5 ms (BF10 = 0.08, d = −0.1), extreme evidence for an indirect task advantage at 25–75 ms (BF10 > 1010, d ≥ 0.2), and extreme evidence for a direct task advantage at 300 ms (BF10 = 1020, d = 2.4).

Methods

  • Six participants (one man, five women; mean age 19.8 years, SD = 1.33) completed a first control session closely modeled on Vorberg et al. (2003), followed by multiple modified sessions with extensive instructions and practice, alternating prime-discrimination and mask-discrimination blocks, and a 300 ms SOA added from session two onward.
  • Analyses used Bayesian single-subject and group-level regression models (logistic for prime discrimination/accuracy; RT models for priming) fitted with brms, with Bayes Factors derived via the Savage-Dickey density ratio; optional stopping continued sessions until strong evidence (BF10 > 10 or < 0.1) was reached in at least four of six SOAs for both tasks.
  • For direct comparison of awareness and priming, mask-discrimination RTs were recoded as accuracy using a median-split technique, then modeled with task and session predictors; reliability and raw-versus-recoded convergence were also assessed.

Limitations

  • Stated: the authors cannot determine which of their changes caused increased prime discrimination; the direct-comparison inference depends on the assumption that the direct task is at least as sensitive as the indirect task; median-split recoding entails possible information loss; and the source of interindividual variability in awareness remains unclear.
  • Stated: LCD versus CRT display differences could not be fully ruled out, though prior work makes this unlikely; the 300 ms SOA may have affected motivation or strategy, although decision criterion did not differ between sessions 1 and 2.
  • Evident from the text: the sample was very small (N = 6) and intensive, group-level models omitted random slopes due to convergence issues, and reliability was poor at the shortest SOA in the recoded mask task (12.5 ms = −0.53) and lower at 300 ms in prime discrimination (0.62), with ceiling performance at 300 ms.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1093/nc/niag052

Cites

40 of 40 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

26 of 35 cited sources checked · 20 supported · 1 flagged · accuracy not yet measured

  • Premisecitation

    Most studies supporting the existence of unconscious priming rely on a dissociation procedure, i.e., showing that prime awareness is not significantly different from chance while also showing a significant priming effect.

    “Most studies supporting the existence of unconscious priming rely on what is usually referred to as a dissociation procedure (Reingold and Merikle 1988, Hannula et al. 2005, Schmidt and Vorberg 2006), i.e. showing that prime awareness is not significantly different from chance, while also showing a significant priming effect.”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites Reingold and Merikle 1988 source

      The paper’s opening discussion (p. 563) says many studies of perception without awareness use a dissociation between an awareness measure and an indirect measure. Its description of the procedure requires null sensitivity on the direct measure and greater-than-zero sensitivity on the indirect measure. It does not specifically establish that most unconscious-priming studies use this procedure, or equate null sensitivity with a result merely not significantly different from chance. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites Hannula et al. 2005 source
    • cites Schmidt and Vorberg 2006 source
  • Premisecitation

    The problem of non-significant p-values can be addressed by using either alternative frequentist or Bayesian methods that can provide evidence for the absence of an effect.

    “Fortunately, this problem can be addressed by using either alternative frequentist (Lakens 2017) or Bayesian (Rouder et al. 2009, Dienes 2014) methods that can provide evidence for the absence of an effect.”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites Lakens 2017 source
    • cites Rouder et al. 2009 source
    • cites Dienes 2014 source
  • Premisecitation

    The double t-test approach is fundamentally flawed.

    “However, this approach is fundamentally flawed (Nieuwenhuis et al. 2011, Meyen et al. 2022).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Nieuwenhuis et al. 2011 source
    • cites Meyen et al. 2022 source
  • Premisecitation

    Non-significant p-values are not sufficient evidence to claim the absence of an effect, as they may also reflect insufficient power to detect a small non-zero effect.

    “Although this is common practice, non-significant p-values are not sufficient evidence to claim the absence of an effect, as they may also reflect insufficient power to detect a small non-zero effect (Keysers et al. 2020).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Keysers et al. 2020 source
  • Study design and conductcitation

    The study tested at the single subject level using a Bayesian statistical framework, instead of performing a frequentist fixed effect analysis at the group level as Vorberg et al. (2003) did.

    “(iv) As in Vorberg et al. (2003), we collected the data of six participants across multiple sessions. However, instead of performing a frequentist fixed effect analysis at the group level as Vorberg et al. (2003) did, we tested at the single subject level using a Bayesian statistical framework.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

      The Subjects passages report six participants tested across multiple sessions in each experiment. The Statistical methods passage describes frequentist group-level tests, including summed χ² statistics and repeated-measures ANOVAs, but it also says performance was tested for significance separately for each subject and condition. Thus the contrast with single-subject testing is overstated; the paper does not explicitly describe its analysis as fixed-effect. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • Classified as probably not a substantial claim. 34
  • May combine two claims. 25
  • The quote only partly states this claim. 1

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    Participants were thoroughly instructed and practised each task before completing the experiment, in contrast to Vorberg et al. (2003), where participants received minimal instructions and did not practice the tasks.

    “(i) Participants were thoroughly instructed and practised each task before completing the experiment, in contrast to Vorberg et al. (2003), where participants received minimal instructions and did not practice the tasks.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

      The Method/Design passage (page unavailable) describes task-specific instructions, including response speed and timing, rather than establishing that participants received minimal instructions. It also reports repeated task sessions; for Experiment 2, data were reported only from sessions three to seven. The paper does not explicitly identify dedicated practice trials, but its account is contrary to the unqualified claim that participants did not practise the tasks. (LLM, conf=0.72)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    In Vorberg et al. (2003), observers were slower to identify the direction of an arrow mask when it followed an arrow prime pointing in the opposite direction, despite being unable to identify the prime direction.

    “In a classic study by Vorberg, D., Mattler, U., Heinecke, A., Schmidt, T., & Schwarzbach, J. (Different time courses for visual perception and action priming. Proceedings of the National Academy of Sciences 2003, 100(10), 6275–6280.), observers were slower to identify the direction of an arrow mask when it followed an arrow prime pointing in the opposite direction, despite being unable to identify”
    • Also stated in: § Introduction · plain
      “For example, in a classic study by Vorberg et al. (2003), observers were faster to identify the direction of an arrow-shaped mask when it followed an arrow-shaped prime pointing in the same (rather than the opposite) direction (priming effect), despite them not being able to discriminate the direction of the prime”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    The decision to collect data until four out of six prime-mask SOAs met the stopping rule was made in order to keep data collection manageable, given that larger samples are often needed to obtain the same strength of evidence for the null than for the alternative hypothesis.

    “The decision to collect data until four out of six prime-mask SOAs met the stopping rule was made in order to keep data collection manageable, given that larger samples are often needed to obtain the same strength of evidence for the null than for the alternative hypothesis (Stefan et al. 2019).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Stefan et al. 2019 source

  • Classified as probably not a substantial claim.

    Premisecitation

    Based on Vorberg et al. (2003), the authors expected most SOAs in the prime discrimination task to show null or near-zero effects.

    “Moreover, based on (Vorberg et al. 2003) we expected most SOAs in the prime discrimination task to show null or near-zero effects.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. 2003 source

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    The stimuli were reconstructed to be as close as possible to the original based on the visual material and specifications in the methods of Vorberg et al. (2003).

    “Based on the visual material and specifications in the methods of Vorberg et al. (2003), we reconstructed the stimuli to be as close as possible to the original.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

  • May combine two claims.

    Premisecitation

    Selecting subjects based on the same data used to demonstrate the absence of awareness is comparable to double-dipping and likely to confound the results with regression to the mean.

    “This practice is problematic because selecting subjects based on the same data used to demonstrate the absence of awareness is comparable to “double-dipping” (Kriegeskorte et al. 2009) and likely to confound the results with regression to the mean (Shanks 2017, Stein et al. 2024).”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites Kriegeskorte et al. 2009 source
    • cites Shanks 2017 source
    • cites Stein et al. 2024 source

  • Classified as probably not a substantial claim.

    Premisecitation

    Many unconscious priming studies, including Vorberg et al. (2003), rely on the double t-test approach.

    “Further, many unconscious priming studies, including Vorberg et al. (2003), rely on what is known as the double t-test approach.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

      In Experiment 1, the Results report priming despite prime-orientation recognition remaining at chance, consistent with comparing an indirect priming measure against a direct awareness measure. But the Statistical methods section says recognition was tested with chi-square tests and reaction-time differences with ANOVAs, not two t-tests. The paper therefore supports the comparison underlying the claim, but not its literal description as a double t-test. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Instead of performing one task per session, the study alternated prime discrimination blocks and mask discrimination blocks in the same session to control for session-level confounds, such as differences in motivation or fatigue.

    “(ii) Instead of performing one task per session, as in Vorberg et al. (2003), we alternated prime discrimination blocks and mask discrimination blocks in the same session (see Fig. 1C) to control for session-level confounds, such as differences in motivation or fatigue.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

      The Experiment 1 Design passage assigns several tasks to separate sessions, supporting the comparison in part. But the Experiment 2 Design passage says choice reaction and prime recognition were administered in an a-d-a order within each session. Thus, 'one task per session' does not describe the cited paper as a whole; the citing authors’ stated reason for their own design is not attributed to Vorberg et al. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S2 (checkable).

    Premisecitation

    In a proper statistical framework, priming and prime awareness should be compared directly on the same scale and within the same statistical test.

    “Instead, in a proper statistical framework, priming and prime awareness should be compared directly on the same scale and within the same statistical test (Meyen et al. 2022, Stein et al. 2024).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Meyen et al. 2022 source
    • cites Stein et al. 2024 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    In the first session, the authors tried to replicate the core original experiment by Vorberg et al. (2003) as closely as possible.

    “In the first session, we tried to replicate the core original experiment by Vorberg et al. (2003) as closely as possible (albeit with slightly different presentation times and different display technology; see Stimuli, timing and apparatus above).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

  • May combine two claims.

    Findingown data

    The mask discrimination results closely resembled the original results by Vorberg et al. (2003): RTs were, on average, longer for incongruent compared with congruent trials, and this priming effect difference became more pronounced as the SOA increased.

    “The mask discrimination results closely resembled the original results by Vorberg et al. (2003): RTs were, on average, longer for incongruent compared with congruent trials (see 3B), and this priming effect difference became more pronounced as the SOA increased (12.5 ms: BF10 = 0.009, d = 0.1; 25 ms: BF10 = 1019, d = 0.8; 37.5 ms: BF10 = 1013, d = 1.4; 50 ms: BF10 = 1013, d = 1.7; 62.5 ms: BF10 =”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

  • The quote only partly states this claim.

    Premisereasoning

    Recoding RT data of the mask discrimination task as accuracy using the median-split technique allowed direct testing of prime discrimination and priming effect within the same scale and statistical test.

    “we recoded the RT data of the mask discrimination task as accuracy using the median-split technique”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Schmidt 2002 source
    • cites Meyen et al. 2022 source

  • May combine two claims.

    Findingown data

    At the group level, participants were able to discriminate the direction of the prime above chance across all SOAs, and this ability strongly increased with SOA.

    “Unlike Vorberg et al. (2003), at the group level, participants were able to discriminate the direction of the prime above chance across all SOAs, and unlike in the original study this ability also strongly increased with SOA (12.5 ms: BF10 = 107, d = 0.2; 25 ms: BF10 = 1015, d = 0.2; 37.5 ms: BF10 > 1040, d = 0.3; 50 ms: BF10 = 1014, d = 0.6; 62.5 ms: BF10 > 1040, d = 0.8; 75 ms: BF10 = 1028, d =”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Vorberg et al. (2003) source

      In the paper’s Experiment 1 Results (indexed as Abstract; no page given), prime-orientation recognition remained at chance across the tested SOAs, supporting the contrast with above-chance discrimination. But Experiment 2 Results and Discussion (also indexed as Abstract; no page given) reports that recognition of short primes improved as the mask occurred later. The claim that recognition did not increase with SOA in the original study is therefore too broad. (LLM, conf=0.85)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingnone given

    All participants were able to successfully complete the practice requirements.

    “All participants were able to successfully complete the practice requirements.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the modified version of the task, prime discrimination performance at the group level was overall above chance and increased at longer SOAs, similar to the priming effect, albeit with smaller effect sizes.

    “In contrast to Vorberg et al. (2003), prime discrimination performance at the group level was overall above chance and increased at longer SOAs, similar to the priming effect, albeit with smaller effect sizes.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Only at the longest SOAs there was more variance in the priming effect than in the prime discrimination task, reflecting ceiling performance in the prime discrimination task at the 300 ms SOA.

    “Only at the longest SOAs there was more variance in the priming effect than in the prime discrimination task, reflecting ceiling performance in the prime discrimination task at the 300 ms SOA.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Most participants performed at chance or near-chance at the two shortest SOAs (12.5 and 25 ms) in the prime discrimination task, with performance improving at longer SOAs.

    “Further, most participants performed at chance or near-chance at the two shortest SOAs, with their performance improving at longer SOAs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the prime detection task, the false alarm rate was similar across SOAs (~4%) and the hit-rate went from roughly 50% to 90% across SOAs, suggesting participants could detect the prime across all SOAs.

    “In our version of this task the false alarm rate was similar across SOAs (~4%) and the hit-rate went from roughly 50% to 90% across SOAs, suggesting participants could detect the prime across all SOAs (see Supplementary Fig. S2).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    All participants showed the same pattern of results in the mask discrimination task: no priming effect at the shortest SOA, with a gradual increase in priming at longer SOAs.

    “all participants showed the same pattern of results in the mask discrimination task, no priming effect at the shortest SOA, with a gradual increase in priming at longer SOAs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Both subject- and group-level results offer strong evidence for an indirect task advantage across all subjects and multiple SOAs.

    “both subject- and group-level results offer strong evidence for an indirect task advantage across all subjects and multiple SOAs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Three participants met the original Vorberg-style criterion for unconscious priming in at least one SOA: participant 3 at 25, 37.5 and 50 ms; participant 4 at 25 ms; participant 5 at 25 and 37.5 ms.

    “Participant 3 at the 25, 37.5 and 50 ms SOAs, participant 4 at the 25 ms SOA, and participant 5 at the 25 and 37.5 ms SOAs”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Even at the longest SOA, the effect size did not exceed 0.32, suggesting that participants performed barely above chance.

    “Note, however, that even at the longest SOA, the effect size did not exceed 0.32, suggesting that participants performed barely above chance.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The SOA-priming profile was highly consistent across subjects, whereas prime awareness was much more variable.

    “the SOA-priming profile was highly consistent across subjects, whereas prime awareness was much more variable.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    At the 300 ms SOA, there was extreme evidence for a larger effect on prime discrimination compared to priming, and this was true for all participants.

    “At the 300 ms SOA there was extreme evidence for a larger effect on prime discrimination compared to priming, and this was true for all participants.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    Analyses were conducted at the single-subject level using a Bayesian sequential testing framework and a pre-defined stopping rule.

    “Analyses were conducted at the single-subject level using a Bayesian sequential testing framework and a pre-defined stopping rule.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Premisenone given

    Some studies may remove participants performing above chance to then test priming on a post-hoc selected subsample.

    “Note also that some studies may remove participants performing above chance to then test priming on a post-hoc selected subsample.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductreasoning

    From session two onwards, a condition with a long SOA of 300 ms was added to reduce the overall difficulty of the experiment and to provide some trials in which the relevant feature of the stimulus was clearly visible.

    “From session two onwards, we added a condition with a long SOA of 300 ms to reduce the overall difficulty of the experiment and to provide some trials in which the relevant feature of the stimulus was clearly visible (see General Procedure for the full rationale).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Awareness could be present in both cases but not reflected in participants' prime discrimination responses.

    “Awareness could be present in both cases but not reflected in their prime discrimination responses.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisenone given

    Many studies have claimed the existence of unconscious priming, but a valid statistical procedure for substantiating such claims is often lacking.

    “Many studies have claimed the existence of unconscious priming, but a valid statistical procedure for substantiating such claims is often lacking.”
    • Also stated in: § Introduction · plain
      “Although many studies have claimed such unconscious priming effects, the topic remains controversial, as claims of unconscious priming often rely on statistical tests that fail to adequately demonstrate the absence of awareness.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Priming and awareness are typically tested by comparing two separate tests.

    “priming and awareness are typically tested by comparing two separate tests”
    • Also stated in: § Introduction · boosted
      “many unconscious priming studies, including Vorberg et al. (2003), rely on what is known as the double t-test approach. In this approach, the comparison of two experimental effects is performed by executing two separate t-tests and independently establishing significance for each.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Nieuwenhuis et al. 2011 source

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    To ensure participants were executing the prime discrimination as intended, the authors devised a progressively difficult practice block in which participants could practice experiment-like trials while also having to perform perfectly on easy trials.

    “To ensure participants were executing the prime discrimination as intended, we devised a progressively difficult practice block in which participants could practice experiment-like trials (12.5–75 ms SOA) while also having to perform perfectly on easy trials (300 ms SOA).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    The study used a Bayesian methodological-statistical framework to test priming effects and prime awareness at the single-subject level across multiple sessions.

    “First, we used a Bayesian methodological-statistical framework to test priming effects and prime awareness at the single-subject level across multiple sessions.”
    • Also stated in: § Introduction · plain
      “we attempted to replicate the study by Vorberg et al. (2003) using a Bayesian framework to test for unconscious priming at the single subject level across multiple experimental sessions.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Participants received extensive training in the prime awareness task to ensure they understood and executed the task as intended.

    “Second, participants received extensive training in the prime awareness task to ensure they understood and executed the task as intended.”
    • Also stated in: § Experiment 1 · plain
      “In session two, participants received more detailed instructions than in session one and completed an extensive practice block for each task before starting the experiment.”
    • Also stated in: § Experiment 1 · plain
      “To ensure participants were executing the prime discrimination as intended, we devised a progressively difficult practice block in which participants could practice experiment-like trials (12.5–75 ms SOA) while also having to perform perfectly on easy trials (300 ms SOA).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    In session two, participants received more detailed instructions than in session one and completed an extensive practice block for each task before starting the experiment.

    “In session two, participants received more detailed instructions than in session one and completed an extensive practice block for each task before starting the experiment.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisenone given

    Typically, researchers ask participants to detect or discriminate a masked prime across many trials and then calculate an aggregated measure of performance, such as proportion correct or Signal Detection Theory's d'.

    “Typically, researchers ask participants to detect or discriminate a masked prime across many trials and then calculate an aggregated measure of performance, such as proportion correct or Signal Detection Theory’s d’.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    No participants were excluded from analysis.

    “No participants were excluded from analysis.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Instead of testing a fixed number of sessions, each participant completed consecutive sessions until the Bayesian statistical framework indicated sufficient evidence for or against unconscious priming based on a pre-defined stopping rule.

    “Further, instead of testing a fixed number of sessions, each participant completed consecutive sessions until our Bayesian statistical framework indicated that there was sufficient evidence for or against unconscious priming based on a pre-defined stopping rule.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    In the double t-test approach, the comparison of two experimental effects is performed by executing two separate t-tests and independently establishing significance for each; when one is significant and the other is not, one concludes they are different.

    “In this approach, the comparison of two experimental effects is performed by executing two separate t-tests and independently establishing significance for each. When one turns out to be significant and the other is not, one concludes that they are different (Nieuwenhuis et al. 2011).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Nieuwenhuis et al. 2011 source

  • Classified as probably not a substantial claim.

    Study design and conductown data

    The study directly compared prime awareness and priming.

    “Third, we directly compared prime awareness and priming.”
    • Also stated in: § Experiment 1 · plain
      “To directly test prime discrimination and priming, we fitted a logistic regression model to the prime discrimination and RT-recoded mask discrimination trial-level data separately for each participant and each SOA.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    All experimental procedures were approved by the Ethical Review Board (VCWE) of the Free University Amsterdam (ref. number VCWE-2021-173).

    “All experimental procedures were approved by the Ethical Review Board (VCWE) of the Free University Amsterdam (ref. number VCWE-2021-173).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    From session three onwards, participants received instructions and completed a 30-trial warm-up block for each task before starting the experiment.

    “From session three onwards, participants received instructions and completed a 30-trial warm-up block for each task before starting the experiment.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The indirect-task advantage is consistent with unconscious priming, under the assumption that the awareness task was at least as sensitive as the priming task.

    “consistent with unconscious priming at short prime-mask stimulus onset asynchronies (SOAs) in three out of six participants, under the assumption that the awareness task was at least as sensitive as the priming task.”
    • Also stated in: § Abstract · boosted
      “provides robust evidence for an indirect-task advantage, consistent with unconscious priming under the assumption that the awareness task was at least as sensitive as the priming task.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The awareness task was at least as sensitive as the priming task.

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    Awareness is frequently assessed by asking participants to complete a difficult task with little or no practice.

    “Another often overlooked issue is that awareness is frequently assessed by asking participants to complete a difficult task (e.g. discriminating briefly presented masked stimuli) with little or no practice.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    Even if participants are trained and know in principle how to perform the task, some may still give up on awareness trials if the task is so difficult that they feel they cannot perform the task meaningfully.

    “Additionally, even if participants are trained and know in principle how to perform the task, some of them may still give up on awareness trials if the task is so difficult that they feel they cannot perform the task meaningfully (i.e. performing above chance).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisenone given

    Performance is often quite low and not statistically distinguishable from chance (P > .05), which is then interpreted as evidence for absence of awareness.

    “The performance is often quite low and not statistically distinguishable from chance (i.e. P > .05), which is then interpreted as evidence for absence of awareness.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    To reduce the chances of participants giving up, the authors included a set of trials with a long SOA (300 ms), in which the relevant feature of the prime stimulus was clearly visible.

    “To reduce the chances of participants giving up, we included a set of trials with a long SOA (300 ms), in which the relevant feature of the prime stimulus participants was clearly visible.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    After session one, each participant completed consecutive experimental sessions until there was strong evidence for either an effect or no effect in at least four out of six prime-mask SOAs both in the prime discrimination and in the mask discrimination task.

    “After session one, each participant completed consecutive experimental sessions until there was strong evidence for either an effect (BF10 10) or no effect (BF10 0.1) in at least four out of six prime-mask stimulus onset asynchronies (SOAs) (see Stimuli and timing) both in the prime discrimination and in the mask discrimination task (see Analysis for a detailed description of the prime and mask”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Only sessions two and onwards were included in the analysis relevant for the application of the optional stopping rule.

    “Only sessions two and onwards were included in the analysis relevant for the application of the optional stopping rule.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisereasoning

    In this context, it is not straightforward to judge whether a participant is performing at chance due to the difficulty of the task or because they do not know how to perform the task, even though the relevant information to perform the task is still consciously perceived.

    “In this context, it is not straightforward to judge whether a participant is performing at chance due to the difficulty of the task or because they do not know how to perform the task, even though the relevant information to perform the task is still consciously perceived.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationreasoning

    Apparent lack of awareness may partially reflect disengagement and/or an inability to convert perceptual experiences into behavioural responses.

    “otherwise, apparent lack of awareness may partially reflect disengagement and/or an inability to convert perceptual experiences into behavioural responses.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The study's findings provide evidence for unconscious priming at the single-subject level using the direct-comparison approach, while the conclusion depends on the assumption that the direct task is at least as sensitive as the indirect task.

    “To conclude, our findings provide evidence for unconscious priming at the single-subject level using the direct-comparison approach, while also emphasizing that this conclusion depends on the assumption that the direct task is at least as sensitive as the indirect task.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The direct task is at least as sensitive as the indirect task to the relevant prime information.

  • May combine two claims.

    Interpretationreasoning

    The variability observed in prime awareness is unlikely to reflect measurement noise alone and is at least partly consistent with genuine individual differences in perceptual sensitivity.

    “Therefore, the variability we observe is unlikely to reflect measurement noise alone and is at least partly consistent with genuine individual differences in perceptual sensitivity”
    Quote in the paperSubstantial claim
52 of 52

  • The authors state this as an assumption.

    The awareness task was at least as sensitive as the priming task.

    “under the assumption that the awareness task was at least as sensitive as the priming task”
    “under the assumption that the awareness task was at least as sensitive as the priming task”

  • The authors state this as an assumption.

    The direct task is at least as sensitive as the indirect task to the relevant prime information.

    “this inference relies on the assumption that the direct task is at least as sensitive as the indirect task to the relevant prime information”
2 of 2

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Evidence at the shortest SOAs is mislabeled as very strong

    Minor

    The largest evidence for chance performance among the three listed SOAs is BF10 = 0.035, equivalent to BF01 ≈ 28.6. By the paper's own thresholds, that is strong-not very strong-evidence.

1 of 1

Cites

Flagged first, then by citations

Cited by

Still being checked.

Related

Authors
Sánchez-Fuenzalida, N., van Gaal, S., van den Hurk, Z., Stein, T., & Fahrenfort, J. J.
Affiliations
  • Vrije Universiteit Amsterdam (NL)
  • Department of Applied and Experimental Psychology, Free University Amsterdam , Van der Boechorststraat 7, 1081 BT Amsterdam ,
  • University of Amsterdam (NL)
Corresponding author
Sánchez-Fuenzalida, Nicolás
ORCID
Venue
Neuroscience of Consciousness
Publisher
Oxford University Press (OUP)
Published
2026
Volume
2026
Issue
1
Pages
niag052-niag052
ISSN
2057-2107 (ISSN-L)
DOI
10.1093/nc/niag052
PMC
PMC13492662
OpenAlex
W4415928735
Semantic Scholar
45d979f09b290c46de89d7fb263bcdaa60b2d6e5
Cited by
0
References
40
FWCI
0.00
Type
journal article
Language
en
Open access
Licence
Locations
  • Neuroscience of Consciousness (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
  • UvA-DARE (University of Amsterdam) (submittedVersion, cc-by)
Funders
  • Templeton World Charity Foundation (30260)
  • European Research Council
  • European Union’s Horizon Europe research and innovation programme (101044686)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural and Behavioral Psychology Studies
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Visual perception and processing mechanisms
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Free Will and Agency
Keywords
unconscious priming, stimulus onset asynchrony, interindividual variability, awareness training

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-11-03.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-11-03.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review