Content-Based Attention Requires Consciousness To Improve Localization

C

ONTENT

-B

ASED

A

TTENTION

R

EQUIRES

C

ONSCIOUSNESS

T

O

I

MPROVE

L

OCALIZATION

BACHELOR THESIS PSYCHOBIOLOGY

NAME OF STUDENT: MUS LEIJTENS

S

ID:

11655224

NAME OF SUPERVISOR: TIMO STEIN

DATE OF SUBMISSION: MAY 22

C

ONTENT

-B

ASED

A

TTENTION

R

EQUIRES

C

ONSCIOUSNESS

T

O

I

MPROVE

L

OCALIZATION

A

BSTRACT

The extent to which cognitive processing is performed unconsciously remains highly debated. While some state that nearly all cognitive processes could happen unconsciously, others state that unconscious processing is very limited. Using various methods researchers have stated that seemingly invisible cues can improve detection. The improvement is due to the unconscious processing of this cue. The current study investigates if cueing can influence the localization of targets when people are unaware of the cue validity. This is done through backward masking, a localization task, a discrimination task, and varying presentation times to ensure unconsciousness in participants. This resulted in one presentation time where participants were unaware of cue validity and were not affected by the cue validity in their localization. Results suggest that if one is unaware of the cue validity, it has no influence on the localization of a target. Hence, content-based attention requires consciousness in order to improve detection and localization. This conclusion provides new insight into the capacity of unconscious processing. Moreover, the awareness measurement in this study could prove useful in the replication of other studies into unconsciousness.

Key words: unconscious perception, cue validity, backwards masking, content-based attention, awareness

I

NTRODUCTION

As soon as one’s eyes are opened in the morning there are many sensory stimuli. To avoid too many stimuli requiring attention at once, it is necessary that the brain prioritizes relevant stimuli, using top-down attention. This ensures that objects do not need to be processed repetitively, which decreases the computational load for the brain and helps with interpreting visual stimuli. Attention can be differentiated in multiple types that focus on different aspects. Feature-based attention focuses on only one feature of a stimulus, such as colour (Baylis & Driver, 1992). Another form of top-down attention is content-based attention. Content-based attention focuses on the nonspatial aspects of objects, these aspects can be of varying complexity (Battistoni, Stein, & Peelen, 2017).

Research into feature-based attention and content-based attention shows that they influence and enhance the recognition and facilitation of objects that are congruent with the expectation (Kravitz & Behrmann, 2011). In previous research it has been found that the visual input for content-specific objects is strengthened when accurate information is given about an object. Therefore, object detection improves for objects that receive content-based attention (Norman, Heywood, & Kentridge, 2013). Enhanced perceptual performance through

unconscious initial stimulus processing is a proposed mechanism for these results (Meijs, Slagter, de Lange, & van Gaal, 2018). Unconscious enhancement may be a mechanism for the effect of content-based attention. The unconscious processing would then lead to the faster detection of expected stimuli. Moreover, this would show that unconscious processing is capable of processing certain shapes to improve the detection of these objects. The complexity of target objects can vary greatly, so where sometimes feature-based attention is sufficient to detect and identify an object, more complicated objects require content-based attention. For example, if the cue is red and there is one red object and one blue, feature-based attention is sufficient to identify the target object. However, is unconscious processing extensive enough to process a complete object or is it quite limited?

The extent of unconscious processing remains highly debated, despite extensive research in this field. Current research is inconclusive about the interpretation of results: how to establish true unconsciousness in participants and which conclusions can be drawn on the topic. Researchers reach different conclusions based on different definitions of consciousness and varying testing methods (Stein, 2019). Due to the discrepancy in theories about consciousness, some researchers argue that nearly all cognitive processes could be unconscious (Hassin, 2013) while others argue that current research shows that the capacity of unconscious processing is in fact restricted (Hesselmann & Moors, 2015). Unconscious processing is most commonly defined as a process where object awareness is absent while object processing is present (to some degree) (Lamy, Salti, & Bar-Haim, 2008).

Abundant studies have looked into the effect of (valid) cues on detection. Using different methods, most came to a comparable conclusion: (content-based) attention improves the detection of stimuli. For instance, using words semantically related or word-parts related to the cue proved to enhance the detection through masking (Costello, Jiang, Baartman, McGlennen, & He, 2009). Likewise, expected visual stimuli (e.g. houses and faces) break through continuous flash suppression (CFS) faster than unexpected or neutral stimuli. This is the case for detection and identification of expected stimuli (Pinto, van Gaal, de Lange, Lamme, & Seth, 2015). In a similar study CFS was also used, but this time combined with an auditive word-representation (Ostarek & Huettig, 2017). This study proved that pictures congruent with the sound were more likely to be detected than noncongruent pictures.

Using electroencephalogram (EEG) analysis, research was conducted into (content-based) attention. A positive peak 80-120 ms after stimulus onset (P1) and a negative peak 160-200 ms (N1) are highly associated with attention (Slagter, Prinssen, Reteig, & Mazaheri, 2016). Cueing has early neurophysiological effects on top down attention (both feature-based and content-based. When participants are cued to a target and an EEG signal is recorded, the P1 and N1 are modulated (Luck, Woodman, & Vogel, 2000; Valdés-Sosa, Bobes, Rodriguez, & Pinilla, 1998). The ERP’s of consciousness are said to be the P1, a visual awareness negativity (VAN) ±200 ms after stimulus onset and a positive peak 300 ms after stimulus onset (P3) (Railo, Koivisto, & Revonsuo, 2011). The P1 is associated with the first visual awareness of a stimuli,

however multiple studies could not associate the P1 component with consciousness (Koivisto & Revonsuo, 2008; Lamy et al., 2008; Sergent, Baillet, & Dehaene, 2005). The VAN corresponds with the successful attentional selection of an object. Lastly, the P3 component is said to reflect the post perceptual actions, which would suggest it is visible after conscious perception. If content-based attention is active 100 ms after stimulus onset, as suggested by the modulated P1 and N1, and visual stimulus awareness (or consciousness) after more than 200 ms, as suggested by the VAN this could indicate that cueing can modify content-based attention without consciousness.

However, all these studies did not disclose how information that should be invisible was still able to interact with the neural networks activated by the cue word. This problem was proposed in a review by Chica & Bartolomeo (2012) on the attentional routes to conscious perception and concluded that a better definition and measurement of consciousness would improve the research into the relationship between attention and consciousness. This measurement helps to solve how invisible information can interact with the neural networks activated by the cue word. Moreover, they report that an objective measure of consciousness is necessary in this field of research since consciousness is currently measured through verbal report. Research by Gayet, Paffen, & Van der Stigchel (2013) suggested that the conscious state can influence visual processing through the selection of visual information that is not yet available for conscious report. Their hypothesis is that the effective threshold for visual input to become conscious is reduced when the target is preactivated through the cue (Gayet, van Maanen, Heilbron, Paffen, & Van der Stigchel, 2016). Their research used the report of participants as a measurement of awareness. The question is, does the absence of conscious report equal the absence of conscious awareness?

To fill this gap in research in current study an extra awareness measure will be used. The current experiment is a modified version of a previous study (Stein & Peelen, 2015) that concluded that object detectability is influenced by content-based expectations. In this previous version a four alternative forced choice (4AFC) localization task was used combined with forward and backward masking. This new experiment tries to replicate the findings of Stein & Peelen with a task consisting of a two alternative forced choice (2AFC), a localization task, a discrimination task and the use of only backward masking. The current study intends to provide new information about the capability of unconscious processing through the addition of the awareness test.

Therefore, in the present study the question that will be answered is: how do cues affect the localization of a target when the participant is unaware of the cue validity? This question will be answered with the use of a backward masking paradigm (Raab, 1963). Whether this process happened unconsciously depends on the participants report of cue validity. Participants who report cue validity above chance level would have to be consciously aware of the stimulus. Vice versa, participants who report at or beneath chance level, were not consciously aware of the stimulus (Schmidt & Vorberg, 2006). Therefore, to determine whether

stimulus localization really happens unconsciously, it is necessary that the cue validity cannot be discriminated above chance level (Stein, 2019).

The current study is expected to replicate previous findings that the detection of validly cued objects will be improved compared to invalidly cued objects (Forder, Taylor, Mankin, Scott, & Franklin, 2016; Hung & Hsieh, 2015; Lupyan & Ward, 2013; Pinto et al., 2015; Stein & Peelen, 2015). D’ or sensitivity index is a measurement for the difference between the hit rate and false alarm rate and therefore provides an indication of detection of a signal. Different stimulation settings are used to aim for a setting that produces a discrimination-d’ that is not significantly different from zero. In this setting the key question of the influence of valid and invalid cueing on localization will be tested. The hypothesis is that in the perfect presentation time, the images are rendered invisible through backward masking, so that participants cannot tell whether or not the word cue was valid, hence discrimination-d’ is not significantly different from zero. Moreover, the localization-d’ is expected to differ significantly when the targets are cued validly compared to invalidly cued targets. This would indicate that the cues still unconsciously improved the localization of valid targets.

M

ETHOD

P

All participants were students at the University of Amsterdam and recruited through the subject pool. Participants received research credit for their participation. This resulted in 95 participants, 17 males and 78 females, aged between 18 and 38 (M: 19.91±2.63). All participants had normal or corrected to normal vision.

SUBJECT EXCLUSION

Participants were submitted to a test run of 10 trials with longer presentation times to ensure their perception was sufficient to participate. In this test run, a participant would be excluded if their accuracy was beneath 90% after more than 3 test runs. None of the participants needed more than 3 test runs, so none were excluded.

APPARATUS AND STIMULI

Stimuli were presented on a 24-inch LCD screen (1920 × 1080 pixels resolution, 120 Hz refresh rate). Presentation times were based on previous research as well as pilot testing (Stein & Peelen, 2015). This resulted in experiment 1A and 1B. In experiment 1A there was a blank period of 8 ms after stimulus presentation (indicated by ‘(8)’ in the rest of this paper), this was absent in experiment 1B. Additionally, in experiment 1A the last presentation time was 58 ms, whereas in experiment 1B the last presentation time was 42 ms. The used presentation times were 8 (8) ms, 17 ms, 25 ms and 58/42 ms. Target stimuli consisted of photographs of eight object categories (car, cat, chair, cup, guitar, hammer, person, shoe). For each of these categories 6 greyscale pictures were selected and randomly picked in the trials (Appendix A). The experiment was built and shown through Matlab using PsychToolBox (Brainard, 1997).

P

There were 768 trials per participant. An example of a trial is shown in figure 1. After reading the information letter and signing the informed consent (Appendix B) participants were verbally instructed that they would start with a fixation period of 500 ms, followed by a word cue that was visible voor 800 ms. After the cue, participants returned to the fixation cross (again 500 ms). Afterwards they would see a picture of one of these object categories and that the picture could be corresponding or noncorresponding with the word cue they were given. In 50% of the time this picture was cued validly. For example, if the word cue was ‘cat’ and the target picture was a cat, the cue was valid. If the word cue was ‘chair’ and the target picture was a hammer, the cue was invalid. The order of validly cued and invalidly cued trials was randomized, as well as the object categories that were shown. The picture was followed by a mask. Participants were instructed that they needed to answer the following questions as accurately as possible and that there was no time-limit on answering. The first question was:

‘Did the target appear left or right?’ This question was for the localization task. The second question was: ‘Was the cue valid or not?’ This question was for the discrimination task. Both questions were answered using arrow keys on the keyboard.

ANALYSIS

For the analysis of the localization task the response ‘left’ (¬) was coded as a hit if the target was presented in the left square, if the response to this target was ‘right’ (®) this was coded as a false alarm. The opposite was the case if the target was presented in the right square. Since the participant can tell where the target appeared by focusing on one side of the screen a correction was applied, by dividing the localization-d’ by the square root of 2 (Tyler & Chen, 2000). For the analysis of the discrimination task the response ‘valid’ (­) was coded as a hit if the cue was valid, if the response to this target was ‘invalid’ (¯) this was coded as a false alarm. The opposite was the case if the cue was invalid. The hit rates and false alarm rates were then used to calculate localization-d’ and discrimination-d’. Using SPSS, a repeated measures ANOVA was used to establish how the presentation time influences discrimination-d’. If the assumption of sphericity was violated, the Greenhouse-Geisser correction was applied. Secondly it was calculated for which presentation time discrimination-d’ was not significantly different from zero, using a one sample t-test. Next, it was determined how the cue validity influences localization-d’ with a repeated measures ANOVA. Lastly, it was calculated for which presentation time the biggest cueing effect presents using a paired t-test.

Figure 1: Example of a trial where the cue was valid, and the target appeared on the left side. First screen shows the fixation cross (500 ms), followed by the cue (800 ms) and another fixation cross (500 ms). After this the target appears, followed by the masks and the questions for the localization and discrimination. The questions remain on screen until they are answered.

R

ESULTS

The goal of the current study was to investigate the effect of cue validity on target detection, when participants are unaware of the cue validity.

EXPERIMENT 1

In figure 2, the results from the discrimination task and the localization task are combined for experiment 1A. The first analysis was a RM-ANOVA to compare the discrimination-d’ against the presentation times. The effect of the presentation time on the discrimination-d’ is displayed in figure 2. As figure 2 shows the discrimination-d’ increased for all presentation times (F(1.80, 79.34) = 592.99, p < 0.001, ηp2 = 0.93).

For every presentation time the discrimination-d’ was significantly different from zero. For the first and shortest presentation time (8 (8) ms) M = 0.08, SD = 0.27, t(44) = 2.08, p = 0.04, for the second presentation time (17 ms) M = 0.35, SD = 0.37, t(44) = 6.19, p < 0.001. For the third presentation time (25 ms) M = 1.04, SD = 0.63, t(44) = 10.97, p < 0.001 and for the last and longest presentation time (58 ms) M = 3.40, SD = 0.90, t(44) = 25.39, p < 0.001.

The second analysis was another RM-ANOVA to investigate the cueing effect on the localization-d’ and the effect of the presentation time on the localization-d’. Figure 2 shows that only for the first three presentation times the cue validity has a positive effect on localization-d’. For the cue validity: F(1, 44) = 32.45, p < 0.001, ηp2 = 0.42, for the presentation

time: F(2.11, 93.01) = 562.08, p < 0.001, ηp2 = 0.93, and for the interaction effect: F(3, 132) = 4.71,

p = 0.004, ηp2 = 0.01. For the fourth presentation time there is no effect of cueing visible in the

localization-d’. The increased presentation time had a positive effect on the localization-d’. The cueing effect was biggest for the third presentation time and smallest for the fourth presentation time. For the first presentation time M = 0.12, SD = 0.27, t(44) = 2.95, p = 0.005. For the second presentation time M = 0.15, SD = 0.32, t(44) = 3.16, p = 0.003. For the third presentation time M = 0.23, SD = 0.33, t(44) = 4.66, p < 0.001. For the fourth presentation time M = -0.01, SD = 0.30, t(44) = -0.30, p = 0.77.

Figure 2: The effect of presentation time on discrimination-d'and presentation time and cue validity on localization-d’ for experiment 1A. First the results for the discrimination-d’. The RM-ANOVA was significant F(1.80, 79.34) = 592.99, p < 0.001. The M±SD discrimination-d for the first presentation time were 0.08±0.27, for the second presentation time 0.35± 0.37, for the third presentation time 1.04±0.63 and for the last presentation time 3.40±0.90. Second the results for the localization-d’. The RM-ANOVA was significant for both the cue validity and presentation time, respectively F(1, 44) = 32.45 and F(2.11, 93.01) = 562.08. The M±SD of the localization-d’ for the four validly cued presentation times were respectively 0.83±0.50, 1.09±0.56, 2.00± 0.72 and 3.03± 0.38. The M±SD of the localization-d’ for the four invalidly cued presentation times were respectively 0.71±0.44, 0.94±0.56, 1.78± 0.73 and 3.04± 0.41. The cueing effect on the localization-d’ was significant for the first three presentation times (p = 0.005, p = 0.003 and p < 0.001). The error bars are SEM.

E

1

:

Experiment 1B was identical to experiment 1A except for the first and last presentation times. In figure 3, the results from the discrimination task and the localization task are combined for experiment 1B. The first analysis was a RM-ANOVA to compare the discrimination-d’ against the presentation times. The effect of the presentation time on the discrimination-d’ is displayed in figure 3. Figure 3 shows that the discrimination-d’ increased with the presentation times. F(1.56, 76.35) = 344.87, p < 0.001, ηp2 = 0.88.

Only for the first presentation time discrimination-d’ was not significantly different from zero. For the first and shortest presentation time (8 ms) M = -0.01, SD = 0.23, t(49) = -0.35, p = 0.73. For the second presentation time (17 ms) M = 0.27, SD = 0.26, t(49) = 7.15, p < 0.001. For the third presentation time (25 ms) M = 0.85, SD = 0.56, t(49) = 10.68, p < 0.001. For the last and longest presentation time (42 ms) M = 2.78, SD = 0.94, t(49) = 20.83, p < 0.001.

The second analysis was another RM-ANOVA to investigate the cueing effect on the localization-d’ and the effect of the presentation time on the localization-d’. Figure 3 shows that only for the first presentation time there is no effect of cue validity on localization-d’. The increasing presentation time had a positive effect on the localization-d’. For the cue validity: F(1, 49) = 20.50, p < 0.001, ηp2 = 0.30. For the presentation time: F(2.34, 114.41) = 639.56, p < 0.001,

ηp2 = 0.93. For the interaction effect: F(3, 147) = 2.35, p = 0.07, ηp2 = 0.05.

The cueing effect was biggest for the third presentation time and smallest for the first presentation time. For the first presentation time M = 0.15, SD = 0.29, t(49) = 0.36, p = 0.72. For the second presentation time M = 0.15, SD = 0.35, t(49) = 2.98, p = 0.004. For the third presentation time M = 0.18, SD = 0.37, t(49) = 3.42, p = 0.001. For the fourth presentation time M = 0.07, SD = 0.34, t(49) = 1.48, p = 0.15.

Figure 3: The effect of presentation time on discrimination-d'and presentation time and cue validity on localization-d’ for experiment 1B. First the results for the discrimination-d’.The RM-ANOVA was significant F(1.56, 76.35) = 344.87, p < 0.001. The M±SD discrimination-d for the first presentation time were 0.01±0.23, for the second presentation time 0.27±0.26, for the third presentation time 0.85±0.56 and for the last presentation time 2.78±0.94. Second the results for the localization-d’. The RM-ANOVA was significant for both the presentation time and cue validity, respectively F(2.34, 114.41) = 639.56 and F(1, 49) = 20.50. The M±SD of the localization-d’ for the four validly cued presentation times were respectively 0.23±0.24, 0.97±0.55, 1.75± 0.67 and 2.84± 0.47. The M±SD of the localization-d’ for the four invalidly cued presentation times were respectively 0.22±0.23, 0.82±0.38, 1.57± 0.62 and 2.77± 0.57. The cueing effect on the localization-d’ was significant for the second and third presentation time (p = 0.004 and p = 0.001). The error bars are SEM.

Lastly, in figure 4 the discrimination-d’ and localization-d’ for every condition are combined, to display the effects of cue validity and presentation time for every presentation time. For the presentation times that were tested in both experiment 1A and experiment 1B, the results are combined in figure 4. While this figure contains no novel information compared to previous figures, it visualizes the results in a different way to improve the comprehension.

Figure 4: The effect of different presentation times on discrimination-d’ and localization-d’. a) Presentation time is 8 ms. Discrimination-d’, localization’d (valid) and localization’d (invalid), respectively M=-0.01, SD = 0.23; M=0.23, SD = 0.24; M = 0.22, SD = 0.23. b) Presentation time is 8 (8) ms, d’s respectively M=-0.08, SD = 0.27; M=0.82, SD = 0.50; M = 0.71, SD = 0.44. c) Presentation time is 17 ms, d’s respectively M=-0.30, SD = 0.32; M=1.02, SD = 0.55; M = 0.87, SD = 0.48. The difference between localization-d’ valid and localization-d’ valid is still significant, t(188) = 1.98 p = 0.049. d) ) Presentation time is 25 ms, d’s respectively M=-0.94, SD = 0.60; M=1.87, SD = 0.70; M = 1.67, SD = 0.68. The difference between localization-d’ valid and localization-d’ valid is still significant, t(188) = 2.01 p = 0.045. e) Presentation time is 42 ms, d’s respectively M=-2.78, SD = 0.94; M=2.84, SD = 0.47; M = 2.77, SD = 0.57. f)

Presentation time is 58 ms, d’s respectively M=-3.40, SD = 0.90; M=3.03, SD = 0.38; M = 3.04, SD = 0.41. All error bars are SEM.

D

ISCUSSION

In the current study the aim was to determine the effect of cue validity on the localization of targets in situations where participants were unaware of cue validity. Previous studies demonstrated that cueing could improve detection, even when participants are not able to report on the cueing. However, it is uncertain whether, when a participant cannot report on cueing, they are truly unaware of the cueing. What remained unclear in previous studies was if the participant was truly unconscious of the cue, which is why the current study came into existence. Through different presentation times, a localization task, a discrimination task and backward masking new findings were provided.

The current results suggest that when a person is unaware of the cue validity, the localization is not influenced by this cue. When looking at the presentation times, there is only one presentation time for which the discrimination-d’ does not significantly differ from zero. This is the 8 ms presentation time of experiment 1B. However, when studying the localization-d’ results for this presentation time, there is no difference between validly cued and invalidly cued targets. For the three following presentation times (8 (8) ms, 17 ms and 25 ms) there is a significant difference between validly cued targets and invalidly cued targets. If the cue was valid, the targets were localized significantly better for these three presentation times. This experiment has replicated the cueing effect that was discussed in plenty previous research; however, it also proves that content-based attention requires consciousness.

The hypotheses stated in the introduction can be partially confirmed and partially rejected. There was one presentation time (8 ms) for which the discrimination-d’ was not significantly different from zero, hence participants were unaware of the cue validity. However, when discrimination-d’ confirmed to the hypothesis, the results from localization-d’ did not. There was no difference visible between validly and invalidly cued targets.

Based on the hypothesis, these results were not as was expected. However, a number of explanations exist in previous research. Over the years, aspects of processing were increasingly believed to happen unconsciously, as discussed in the introduction (Hassin, 2013). However, it was also previously suggested that some processes simply cannot happen unconsciously. For example, it was found that orientation specificity is only present when objects are visible (Kiefer, 2007). This would explain why the localization was not different for validly cued targets when participants were unaware of cue validity, since consciousness was necessary for the localization. In the introduction some neurophysiology was introduced on the topic and suggested that the P1 is not fully associated with conscious access to stimuli, since multiple studies could not find the relationship between the two. However, other studies do provide the relationship between consciousness and the P1 component, therefore it could be that the hypotheses of following studies were confirmed, rather than the studies referenced in the introduction (Kornmeier & Bach, 2005; Mathewson, Gratton, Fabiani, Beck, & Ro, 2009; Roeber et al., 2008). In that case, it could be that the content-based attention and conscious

processing were active at the same time which could explain why there was no cueing effect without cue awareness. Additionally, when there is a goal to be obtained with a target, such as target manipulation, it is also said that conscious access is required (Ansorge, Kunde, & Kiefer, 2014). This could explain the results that were found in the present study. The identification and localization of the invisible target based on the cue could be a goal that cannot be obtained without conscious awareness.

There are also some limitations to the used methods of the current study that could explain some of the results. For example, due to technical limitations, there was no presentation time possible in between the 8 ms and the 17 ms. It might be the case that there is a time frame in between these presentation times for which discrimination-d’ is still not significantly different from zero, while the localization-d’ is improved by the valid cues. Perhaps the fact that cues and instructions were only available in Dutch or English made the task more challenging since there were many participants that did not have either of these languages as their native language.

As mentioned in the introduction, this study was a replication of a previous study (Stein & Peelen, 2015) with an extra awareness measure. This provided new knowledge about the capacity of conscious processing. This new knowledge suggests that the capacity of conscious processing is in fact limited and not all cognitive processes can take place while unconscious. The main aspect of this conclusion is based on the fact that previously known cueing effects were replicated in the current study, however cueing effects were not present when the participants were truly unconscious of the cue validity. For other studies in the field of unconscious processing the use of the awareness measure used in current experiment could confirm whether their results were obtained consciously or unconsciously.

For further research it would be interesting to test the presentation times in between 8 and 17 ms, to see if there might be a presentation time for which participants are still unaware of cue validity but are affected by the cueing. Furthermore, previous experiments that tested cueing in unconscious participants and used the report of target awareness as measurement, could replicate their studies using this awareness measurement to find out if their results were really based on unconscious participants.

In conclusion, this study has used backward masking to establish if cue validity affects target localization when a person is unaware of the cue validity. Based on previous research it was expected that the valid cues would improve the target localization compared to the invalid cues. Using varying presentation times, it was shown that the localization of the target is not influenced by the cue validity when participants are unaware of the cue validity, since the localization-d’ does not differ between validly and invalidly cued targets. While this is not congruent with the expectations based on previous research, it does provide new insight into the capacities of unconscious processing: it suggests that content-based attention requires consciousness. More research is needed in order to establish the connection between CBA and consciousness, but this study provides the first step.

R

EFERENCES

Ansorge, U., Kunde, W., & Kiefer, M. (2014). Unconscious vision and executive control: How unconscious processing and conscious action control interact. Consciousness and Cognition, 27(1), 268–287. https://doi.org/10.1016/j.concog.2014.05.009

Battistoni, E., Stein, T., & Peelen, M. V. (2017). Preparatory attention in visual cortex. Annals of the New York Academy of Sciences, 1396(1), 92–107. https://doi.org/10.1111/nyas.13320 Baylis, G. C., & Driver, J. (1992). Visual parsing and response competition: The effect of

grouping factors. Perception & Psychophysics, 51(2), 145–162. https://doi.org/10.3758/BF03212239

Brainard, D. H. (1997). The Psychophysics Toolbox. Spatial Vision, 10(4), 433–436. https://doi.org/https://doi.org/10.1163/156856897X00357

Chica, A. B., & Bartolomeo, P. (2012). Attentional routes to conscious perception. Frontiers in Psychology, 3(1), 1–12. https://doi.org/10.3389/fpsyg.2012.00001

Costello, P., Jiang, Y., Baartman, B., McGlennen, K., & He, S. (2009). Semantic and subword priming during binocular suppression. Consciousness and Cognition, 18(2), 375–382. https://doi.org/10.1016/j.concog.2009.02.003

Forder, L., Taylor, O., Mankin, H., Scott, R. B., & Franklin, A. (2016). Colour Terms Affect Detection of Colour and Colour-Associated Objects Suppressed from Visual Awareness. PLoS ONE 11(3):, 11(3), 1–19. https://doi.org/10.1371/journal.pone.0152212

Gayet, S., Paffen, C. L. E., & Van der Stigchel, S. (2013). Information matching the content of visual working memory is prioritized for conscious access. Psychological Science, 24(12), 2472–2480. https://doi.org/10.1177/0956797613495882

Gayet, S., van Maanen, L., Heilbron, M., Paffen, C. L. E., & Van der Stigchel, S. (2016). Visual input that matches the content of visual working memory requires less (not faster) evidence sampling to reach conscious access. Journal of Vision, 16(11), 1–20. https://doi.org/10.1167/16.11.26

Hassin, R. R. (2013). Yes It Can: On the Functional Abilities of the Human Unconscious. Perspectives on Psychological Science, 8(2), 195–207. https://doi.org/10.1177/1745691612460684

Hesselmann, G., & Moors, P. (2015). Definitely maybe: can unconscious processes perform the same functions as conscious processes? Frontiers in Psychology, 6(584). https://doi.org/10.3389/fpsyg.2015.00584

Hung, S.-M., & Hsieh, P.-J. (2015). Syntactic Processing in the Absence of Awareness and Semantics. Journal of Experimental Psychology: Human Perception and Performance, 41(5), 1376–1384. https://doi.org/10.1037/xhp0000094

Kiefer, M. (2007). Top-down modulation of unconscious “automatic” processes: A gating framework. Advances in Cognitive Psychology, 3(1–2), 289–306. https://doi.org/10.2478/v10053-008-0031-2

Koivisto, M., & Revonsuo, A. (2008). The role of selective attention in visual awareness of stimulus features: Electrophysiological studies. Cognitive, Affective, & Behavioral Neuroscience, 8(2), 195–210. https://doi.org/10.3758/CABN.8.2.195

Kornmeier, J., & Bach, M. (2005). The Necker cube - An ambiguous figure disambiguated in early visual processing. Vision Research, 45(8), 955–960. https://doi.org/10.1016/j.visres.2004.10.006

Kravitz, D. J., & Behrmann, M. (2011). Space-, object-, and feature-based attention interact to organize visual scenes. Attention, Perception, and Psychophysics, 73(8), 2434–2447. https://doi.org/10.3758/s13414-011-0201-z

Lamy, D., Salti, M., & Bar-Haim, Y. (2008). Neural correlates of subjective awareness and unconscious processing: An ERP study. Journal of Cognitive Neuroscience, 21(7), 1435–1446. https://doi.org/10.1162/jocn.2009.21064

Luck, S. J., Woodman, G. F., & Vogel, E. K. (2000). Event-related potential studies of attention. Trends in Cognitive Sciences, 4(11), 432–440. https://doi.org/10.1016/S1364-6613(00)01545-X Lupyan, G., & Ward, E. J. (2013). Language can boost otherwise unseen objects into visual awareness. Proceedings of the National Academy of Sciences of the United States of America, 110(35), 14196–14201. https://doi.org/10.1073/pnas.1303312110

Mathewson, K. E., Gratton, G., Fabiani, M., Beck, D. M., & Ro, T. (2009). To See or Not to See: Prestimulus alpha Phase Predicts Visual Awareness. The Journal of Neuroscience, 29(9), 2725–2732. https://doi.org/10.1523/JNEUROSCI.3963-08.2009

Meijs, E. L., Slagter, H. A., de Lange, F. P., & van Gaal, S. (2018). Dynamic interactions between top–down expectations and conscious awareness. The Journal of Neuroscience, 38(9), 2318– 2327. https://doi.org/10.1523/JNEUROSCI.1952-17.2017

Norman, L. J., Heywood, C. A., & Kentridge, R. W. (2013). Object-Based Attention Without

Awareness. Psychological Science, 24(6), 836–843.

https://doi.org/10.1177/0956797612461449

Ostarek, M., & Huettig, F. (2017). Spoken Words Can Make the Invisible Visible-Testing the Involvement of Low-Level Visual Representations in Spoken Word Processing. Journal of Experimental Psychology: Human Perception and Performance, 43(3), 499–508. https://doi.org/10.1037/xhp0000313

Pinto, Y., van Gaal, S., de Lange, F. P., Lamme, V. A. F., & Seth, A. K. (2015). Expectations accelerate entry of visual stimuli into awareness. Journal of Vision, 15(8), 1–15. https://doi.org/10.1167/15.8.13

Raab, D. H. (1963). Backward masking. Psychological Bulletin, 60(2), 118–129.

Railo, H., Koivisto, M., & Revonsuo, A. (2011). Tracking the processes behind conscious perception: A review of event-related potential correlates of visual consciousness. Consciousness and Cognition, 20(3), 972–983. https://doi.org/10.1016/j.concog.2011.03.019 Roeber, U., Widmann, A., Trujillo-Barreto, N. J., Herrmann, C. S., O’Shea, R. P., & Schröger, E.

(2008). Early correlates of visual awareness in the human brain: Time and place from event-related brain potentials. Journal of Vision, 8(3), 1–12. https://doi.org/10.1167/8.3.21 Schmidt, T., & Vorberg, D. (2006). Criteria for unconscious cognition: Three types of

dissociation. Perception and Psychophysics, 68(3), 489–504. https://doi.org/10.3758/BF03193692

Sergent, C., Baillet, S., & Dehaene, S. (2005). Timing of the brain events underlying access to consciousness during the attentional blink. Nature Neuroscience, 8(10), 1391–1400. https://doi.org/10.1038/nn1549

Slagter, H. A., Prinssen, S., Reteig, L. C., & Mazaheri, A. (2016). Facilitation and inhibition in attention: Functional dissociation of pre-stimulus alpha activity, P1, and N1 components. NeuroImage, 125, 25–35. https://doi.org/10.1016/j.neuroimage.2015.09.058

Stein, T. (2019). The breaking continuous flash suppression paradigm: Review, evaluation, and outlook. In Transitions between Consciousness and Unconsciousness.

Stein, T., & Peelen, M. V. (2015). Content-Specific Expectations Enhance Stimulus Detectability by Increasing Perceptual Sensitivity Content-Specific Expectations Enhance Stimulus Detectability by Increasing Perceptual Sensitivity. Journal of Experimental Psychology: General. https://doi.org/10.1037/xge0000109

Tyler, C. W., & Chen, C. C. (2000). Signal detection theory in the 2AFC paradigm: Attention, channel uncertainty and probability summation. Vision Research, 40(22), 3121–3144. https://doi.org/10.1016/S0042-6989(00)00157-7

Valdés-Sosa, M., Bobes, M. A., Rodriguez, V., & Pinilla, T. (1998). Switching Attention without Shifting the Spotlight: Object-Based Attentional Modulation of Brain Potentials. Journal of Cognitive Neuroscience, 10(1), 137–151. https://doi.org/10.1162/089892998563743

A

PPENDIX

A

A

PPENDIX

B

Information letter and informed consent

- Content-based attention without awareness -

Dear participant,

First, thank you for your interest! Before the experiment starts, it is important that you are informed about the procedures. Therefore, we would like you to read this information letter carefully. Please do not hesitate to ask for clarification about this text or the general procedure. If anything might be unclear, the researcher will answer your questions.

Goal of the study

The goal of this study is to investigate the ability to process certain features of faces and objects unconsciously.

Procedure

In this study, you will be presented with very briefly shown pictures of so-called oriented gratings or photographs of objects on a computer screen. They will be followed by masks, i.e. by patterns rendering them very difficult to see. You have three very simple tasks: Detection, localization, and discrimination. In the “detection” task, you indicate whether or not you have seen a target grating or object (a picture is present in only 66% of the trials). In the “localization” task, you guess whether the picture was presented left or right from the fixation cross. The “discrimination” task differs between versions of this experiment and the experimenter will inform you which version of the task you are going to do (e.g. indicate whether a word preceding the stimulation sequence matched the presented grating or object photograph). For this task, you will receive additional information from the experimenter after filling out this form. The total duration of the task will be 60 minutes and there will be multiple opportunities for you to take a break.

Voluntary participation

There are no consequences if you decide now not to participate in this study. During the experiment, you are free to stop participating at any moment without giving a reason for doing so.

Discomfort, risks and insurance

As with any research at the University of Amsterdam, a standard liability insurance applies. The UvA is legally obliged to inform the Dutch Tax Authority (“Belastingdienst”) about financial compensation for participants. You may receive a letter from the UvA with a payment overview and information about tax return.

Your privacy is guaranteed

Your personal information (about who you are) remains confidential and will not be shared without your explicit consent. Your research data will be analyzed by the researchers that collected the information. Research data published in scientific journals will be anonymous and cannot be traced back to you as an individual. Completely anonymized data can be shared with other researchers.

Compensation

As compensation for your participation, you receive 1,5 research credits.

Further information

Should you have questions about this study at any given moment, please contact the responsible researcher Timo Stein (timo@timostein.de), Formal complaints about this study can be addressed to the Ethics Review Board; Hans Phaf (R.H.Phaf@uva.nl).Thank you,

C O N S E N T F O R M

In this form, we refer to the information letter describing the research in which you participate. By signing this form, you declare that you understand the nature and methods of this study as described in the information letter.

Should you have questions about this study at any given moment, please contact the responsible researcher Timo Stein (timo@timostein.de). Formal complaints about this study addressed to the Ethics Review Board; Hans Phaf (R.H.Phaf@uva.nl).

Signed in duplicate

[PARTICIPANT]

• I am 16 or older;

• I have read and understand the information letter;

• I agree to participate in this study and I agree with the use of the data that are collected; • I reserve the right to withdraw my participation from the study at any moment without

providing any reason.

... ... participant name participant signature

... date

[RESEARCHER]

• I informed the participant about the research;

• I am willing to answer any possible questions about the research to the best of my ability. ... ...

researcher name researcher signature ...