Taking consciousness for real: Increasing the ecological validity of the study of conscious vs. unconscious processes

Abstract

The study of consciousness has developed well-controlled, rigorous methods for manipulating and measuring consciousness. Yet, in the process, experimental paradigms grew farther away from everyday conscious and unconscious processes, which raises the concern of ecological validity. In this review, we suggest that the field can benefit from adopting a more ecological approach, akin to other fields of cognitive science. There, this approach challenged some existing hypotheses, yielded stronger effects, and enabled new research questions. We argue that such a move is critical for studying consciousness, where experimental paradigms tend to be artificial and small effect sizes are relatively prevalent. We identify three paths for doing so-changing the stimuli and experimental settings, changing the measures, and changing the research questions themselves-and review works that have already started implementing such approaches. While acknowledging the inherent challenges, we call for increasing ecological validity in consciousness studies.

Keywords: augmented reality; consciousness; ecological studies; naturalistic designs; unconscious processing; virtual reality.

Figure 1

Surveys of current methods in the field of consciousness studies On the left (A), a classification of methods used in 263 neuroscientific experiments manipulating the content of consciousness, whose results were interpreted in light of four theories of consciousness (taken from the ConTraSt database⁴¹). On the right (B), a classification of methods used in 387 behavioral experiments that examined semantic processing (n = 277) and attentional capture (n = 110) without awareness (data collected for two unpublished meta-analyses). CFS, continuous flash suppression.

Figure 2

Examples of ecological stimuli (A) Natural scenes with objects substituted to create violations of semantic expectations. Adapted from Shir et al. (B) Biological motion cues presented by point-light walkers, suppressed using CFS, left: timeline of a single trial, right: illustration of male (blue) and female (red) stimuli, differing in their structural and motion properties. Adapted from Faivre and Koch. (C) Vestibular cues facilitating emergence into awareness of congruent suppressed visual motion cues, left: image of a participant sitting in a chair that rotates, right: CFS presentation of motion cues. Adapted from Salomon et al. (D) “Real-life” CFS allows to suppress real, 3D objects from awareness, left: illustration of the setup, top-right: timeline of a single trial, from the point of view of the participant, bottom-right: examples of 3D-printed stimuli depicting objects (left) and their scrambled versions (right). Adapted from Korisky et al. (E) Inattentional blindness in the real world. Left: participants failed to notice a gun in full view on a driver’s dashboard. Adapted from Simons and Schlosser. Right: participants failed to notice a unicycling clown despite crossing their path. Adapted from Hyman et al. (F) A multi-trial inattentional blindness VR paradigm. The participant is riding a bus through a street while following one of three moving bees (circled here for illustration purposes). Billboards in the background and on the bus stations show intact and scrambled stimuli. Adapted from Hirschhorn et al. (G) A demonstration of limited awareness of color in the periphery using immersive 360° videos. Left: visualizations of displays presented to participants in two eccentricities of color desaturation in the periphery, right: awareness to color desaturation across participants. Adapted from Cohen et al. (H) Decoding conscious perception of natural stimuli (movies) in unresponsive patients based on neural responses of healthy participants. Left: movie frames with high between-participants agreement on being suspenseful, right: visualization of selected ICA components from the healthy participants (top row), and their expression in two patients (two bottom rows), showing the degree of synchrony between neural responses of the single patient and the group. Adapted from Naci et al.

Figure 3

Examples of more ecological measures for consciousness studies (A) Eye-tracking measures reveal the effect of binocular rivalry, both by the average OKN speed (top) and the average pupil size (bottom). Adapted from Frässle et al. (B) SCR measurements demonstrate unconscious fear conditioning. Top: experimental paradigm. Participants were presented with visible/invisible fearful faces, either coupled with a shock (CS+) or not (CS−). Bottom: a difference in SCRs between CS+ and CS− was found during unconscious processing in the early (first half, blue) part of the experiment, while for conscious processing, the effect was found during the late (second half, red) part. Adapted from Raio et al. (C) Motion tracking suggests differential reaching trajectories for a condition where the prime and target were congruent (top) as opposed to incongruent (bottom). As can be seen, trajectories in the former case are more distant and direct than in the latter condition. Adapted from Finkbeiner and Friedman. (D) Drawing as a report method reveals recall differences between aphantasic and control groups. Participants were asked to draw an image either from memory or while it was presented. In the figure are examples of drawings made by individual participants (rows) in each condition, demonstrating a differential amount of detail both between conditions and also between aphantasics and controls in the recall condition, especially for images that were more extensively described (high memory). Adapted from. Bainbridge et al.

Figure 4

Ecological studies in consciousness enabling new questions (A) Phenomenal awareness without access. Left: the experimental design in the different conditions of the study. The critical one appears in the second row: an ongoing pink noise is presented alongside multiple sounds, which gradually disappear until only the noise remains. Participants are then asked whether they hear anything. Then, the noise stops, and participants are asked whether they heard the change and are requested to detect which of two sounds were presented. The table in the bottom row shows the different types of trials, based on participants’ responses. Access trials are those where participants noticed the noise when asked. Phenomenal trials are those where they denied hearing the sound in real time but noticed the change. No-consciousness trials are those where they denied hearing the stimulus and did not notice the change. Right: accuracy in retrospectively identifying the background noise for access trials (left), phenomenal trials (middle), and no consciousness (right). Adapted from Amir et al. (B) Detecting brain areas that track the physical stimulus vs. its conscious percept using blinks. Deconvolved high-frequency broadband responses from two patients with intracranial electrodes (columns). Responses shown to differentiate early sites (left sub-columns, green) from high-order sites (right sub-columns, yellow) when the content of consciousness differs from the physical stimulus (voluntary or spontaneous blinks, two bottom rows), but not when gaps are introduced to the visual stimulus itself (top row). Adapted from Golan et al. (C) The diminishing of the readiness potential (RP) for deliberate, non-arbitrary decisions. Middle: time course of deliberate (red shades) and arbitrary (blue shades) easy and hard decisions in electrode Cz; Top inset: mean voltage of the RP for the time bin [−0.5, 0] before response; Bottom inset: typical scalp distributions. Notably, the RP significantly differs from zero and displays a typical scalp distribution for arbitrary decisions only. Adapted from Maoz et al.